The Ohio State University
Knowledge Bank
kb.osu.edu
Ohio State Law Journal (Moritz College of Law)
Ohio State Law Journal: Volume 60, Issue 4 (1999)
1999
Judicial Deference to EEOC Interpretations of
the Civil Rights Act, the ADA, and the ADEA:
Is the EEOC a Second Class Agency?
Wern, Theodore W.
Ohio State Law Journal, vol. 60, no. 4 (1999), 1533-1587.
http://hdl.handle.net/1811/65026
Downloaded from the Knowledge Bank, The Ohio State University's institutional repository
Judicial Deference to EEOC Interpretations of the
Civil Rights Act, the ADA, and the ADEA: Is the
EEOC a Second Class Agency?
THEODORE W. WERN*
This Note respondsto concernsthat the EEOC'sinterpretationsof Title VI,
the ADA, and the ADEA have not received theirfairshare ofjudicial deference.
The authorexamines Supreme Courtopinions in which the Court wasfaced with
a choice of eitheraccepting or rejecting the EEOC's interpretationof one of its
empowering statutes. This examinationproduces an overall "rateof deference "
for EEOC interpretations.The authorcompares the EEOCdeference rate to the
rate of Supreme Courtdeference to all agencies (the "baseline') and concludes
that the EEOC receives considerably less deference than other federal
administrative agencies. Hence, the EEOC can properly be considered a
"secondclass" agency. To test this conclusion, the authorexamines some of the
leadingjudicial deference cases arisingunder each of the EEOC's empowering
statutes. The result is the same. The authorthen explores the reasonswhy EEOC
interpretations of these statutes have received so little judicial deference.
Finally, he offers some modest recommendations to rescue the EEOCfrom its
second class status.
I. INTRODUCTION
The Equal Employment Opportunity Commission (EEOC) was created to
enforce the broad guidelines of the Civil Rights Act of 1964.1 At present the
EEOC has enforcement power over Title VII of the Civil Rights Act 2 Title I of
* J.D. candidate, The Ohio State University College of Law. I wish to thank Professors
Ruth Colker and Arthur Greenbaum for their invaluable assistance in polishing this Note. I also
wish to thank my family for instilling in me the patience and work ethic required to complete
this Note; my dearest Sally Burma for her love and inspiration; and Kaylynn Two Trees for her
precious wisdom. This Note is dedicated to the memory of William Urban.
1 See Civil Rights Act of 1964, Pub. L. No. 88-352, § 705,78 Stat. 241,258-59 (codified
as amended at 42 U.S.C. § 2000e-4 (1994)) (providing for composition, duties, and powers of
the EEOC). For a comprehensive view of the structure and day-to-day operations of the EEOC,
see U.S. COMM'N ON CIVIL RIGHTS, HELPING EMPLOYERS COMPLY WITH THE ADA: AN
ASSESSMENT OF HOW THE UNITED STATES EQUAL EMPLOYMENT OPPORTUNiTY COMMISSION is
ENFORCING TITLE I OF THE AMERICANS WITH DISABnmnES Acr, 38-51 (1998) [hereinafter
CCR REPORT ON ADA TITLE 1].
2See 42 U.S.C. § 2000e-5(a) (1994) ('The Commission is empowered... to prevent any
person from engaging in any unlawful employment practice as set forth in section 2000e-2 or
2000e-3 of this title."); 42 U.S.C. § 2000e-12(a) (1994) ('The Commission shall have authority
from time to time to issue, amend, or rescind suitable procedural regulations to carry out the
provisions of this subchapter.").
1534
OHIOSTATE LAWJOURNAL
[Vol. 60:1533
the Americans with Disabilities Act (ADA),3 and the Age Discrimination in
Employment Act (ADEA).4 Among each of these statutes, a common thread
exists: eradication of employment discrimination. However, the EEOC does not
have the same authority to enforce and interpret all three statutes. For example,
under the Civil Rights Act, Congress did not grant the EEOC the authority to
promulgate rules with the force of law,5 but such a grant is clear under the
ADA.6 With differing grants of authority to the EEOC under different statutes,
and varying views regarding the amount of judicial deference that should be
given to administrative agencies in general, it is no wonder that the standard of
deference for EEOC guidelines is undefined.
With respect to EEOC deference, the most contentious issues arise when
judges confront interpretive guidelines as opposed to legislative-type rules or
regulations. 7 Even if the EEOC has clear statutory authority to promulgate
"regulations" with the force of law (as it does in the ADA context), what happens
when it issues "interpretive guidance" as an appendix to the regulations 8 or
3Title I of the ADA covers discrimination in employment decisions, See Americans with
Disabilities Act of 1990, 42 U.S.C. § 12116 (1990) (authorizing the EEOC to promulgate
regulations implementing Title I of the ADA). Title II (public services) and the transportationrelated sections of Titles II and II (public accommodations) are enforced, respectively by the
Attorney General and Department of Transportation. See id. § 12134(a) (authorizing the
Attorney General); see id. §§ 12149, 12164, 12186 (authorizing the Secretary of the
Transportation).
4 The ADEA was originally enforced by the Wage and Hour Division of the United States
Department of Labor. In 1978, President Carter, acting pursuant to the Reorganization Act of
1977, transferred ADEA enforcement power to the EEOC. See Reorg. Plan No. 1 of 1978, 3
C.F.R. 321 (1978), reprintedin 5 U.S.C. § 2 app. at 1574 (1994) (transferring enforcement and
regulatory power over the ADEA from Department of Labor to the EEOC).
5 The EEOC was only given authority to issue "suitable procedural regulations.' See 42
U.S.C. § 2000e-12(a) (1994). Furthermore, EEOC guidelines enforcing Title VII of the Civil
Rights Act of 1964 were not promulgated pursuant to the Administrative Procedure Act
(APA), 5 U.S.C. § 553 (1994). Such procedural requirements will be discussed in more detail
below, but they essentially require an agency to post its regulations in the Federal Register for
notice and comment before they are published in the Code of Federal Regulations. See 5
U.S.C. § 553 (1998).
6See supranote 3 (discussing who has the authority to enforce various statutes).
7 Legislative-type regulations, unlike interpretive guidelines, cany the force of law, and
thus can only be disregarded in rare circumstances. See KENNETH CULP DAVIS & RICHARD
PIERCF, JR., ADMINisTRATivE LAW TREATISE § 6.3, at 250-55 (3d ed. 1994) (discussing the
binding effect of legislative rules). This distinction is discussed in detail below. See infra Part
II.C.
8 The EEOC promulgated regulations at 29 C.F.R. § 1630 (1998), but also attached an
appendix titled "Interpretive Guidance," which mirrors the regulations in citation form (except
for the addition of "app.' at the end of each cite). In substance, the Interpretive Guidance is
1999]
JUDICIAL DEFERENCETO THEEEOC
1535
"enforcement guidance?" 9 These types of guidance carry less inherent weight
than regulations, but often reflect agency expertise that no other person or entity
can match.1 0
Currently, no settled judicial view exists as to when the courts should defer
to EEOC interpretations, and some commentators argue that the EEOC receives
less than its fair share of deference."1 The purpose of this Note is to determine
whether the EEOC actually receives less deference than other agencies, and if so,
to explore the means of redeeming the agency from its second class status.
The order of analysis is as follows. Part I explores, as background material,
the two major theoretical models for agency deference: "earned" and
"compelled" deference. Part III presents a statistical comparison between overall
Supreme Court deference to agencies and the EEOC. Part IV discusses the major
deference problems arising from EEOC interpretations of Title VII, the ADEA,
and the ADA, with a special focus on the ADA. Part V presents some views on
why the EEOC has received its second class status. Finally, Part VI provides
recommendations which, if implemented, could render the EEOC a more
effective and respectable administrative agency.
quite different in that it seeks to define any ambiguous words or phrases codified in the ADA.
The real question, which will be explored at length in Part IV.C.1 below, is whether the
Guidance should be entitled to the weight of its neighboring "regulations."
9This type of guidance is entitled to the least deference. It is not published for notice and
comment pursuant to the APA. Furthermore, it is often published many years after enactment
of the statute. See Colin S. Diver, Statutory Interpretationin the AdministrativeState, 133 U.
PA. L. REV. 549, 562 n.95 (1985) (citing contemporaneity with enactment as one factor that
increases the weight of an agency interpretation). Enforcement guidance will be discussed in
detail below in the context of the ADA. See infra Part IV.C.2.c.
10
Congress has recognized the unique expertise of the EEOC in the area of employment
discrimination. See H.R. REP. No. 92-238 (1971), reprintedin 1972 U.S.CAA.N 2137,2146
("Administrative tribunals are better equipped to handle the complicated issues involved in
employment discrimination cases.").
11
See Ruth Colker, The Americans with DisabilitiesAct: A Windfall for Defendants, 34
HARV. C.R.-C.L. L. REv. 99, 144-46 (1999) (arguing that EEOC interpretations of civil rights
statutes have not fared well in the courts). But cf, CCR REPORT ON ADA TITLE I, supranote 1,
at 241-69 (criticizing the EEOC's interpretive and enforcement practices in the context of Title
I of the ADA); Jamie A. Yavelberg, Revival of Skidmore v. Swifb Judicial Deference to
Agency InterpretationsAfter EEOC v. ARAMCO, 42 DUKE L.J 166,200 (1992) (arguing that
in the context of EEOC guidance "it is simply not good practice for courts to defer to
interpretive rules" because they are exempt from APA requirements).
1536
OHIO STATE LA WJOURNVAL
[Vol. 60:1533
II. THEORETICAL MODELS OF AGENCY DEFERENCE
The following subsections present two well-settled models of deference to
agency guidance: earned and compelled deference. 12 Earned deference applies to
agency guidance that has not undergone the basic procedures necessary to
formulate a legally binding rule.13 Courts are not compelled to follow this type
of guidance, rather, they conduct their own evaluation to determine the proper
weight of the agency's opinion. 14 On the other hand, compelled deference
applies to rules that have passed all necessary procedures, and can properly be
considered binding. 15 A court is compelled to follow this type of guidance unless
it directly conflicts with the statute that originally empowered the agency. 16 Both
models will be explained in greater detail below.
The basic similarity between the compelled and earned deference models is
that unambiguous statutory language supersedes both models. 17 A court must
first consider whether Congress has spoken clearly to the disputed issue. 18 If so,
then neither deference model applies, and the inquiry ends. 19
12
Scholars have also proposed some creative deference models that the courts have not
adopted. See Thomas W. Merrill, JudicialDeference to Executive Precedent, 101 YALE LJ.
969, 1010-13 (1992) (proposing the "executive precedent model"); John S. Moot, An Analysis
of JudicialDeference to EEOC Interpretive Guidelines, I ADMIN. L.J. 213, 215-16 (1987)
(proposing the "modified deferential model").
13 The procedure required for a binding informal rule is conmonly known as the "notice
and comment" process, and is governed by § 553 of the APA. See 5 U.S.C. § 553 (1994). For a
more detailed analysis of this procedure, see infra note 56.
14
See infra Part IIA (discussing earned deference).
See infra Part II.B (discussing compelled deference).
16
See id.
17 'Me notion that plain statutory meaning precludes consideration of agency guidance is
emphatically announced by Justice Stevens in Chevron, US.A., Inc. v. Natural Resources
Defense Council: "If the intent of Congress is clear, that is the end of the matter, for the court,
as well as the agency, must give effect to the unambiguously expressed intent of Congress."
467 U.S. 837, 842-43 (1984). This is the first step of Chevron's notorious two-step deference
test. The second step refers to situations in which the statute is ambiguous and must be clarified
by agency guidance. See infra note 51 and accompanying text (describing the second step of
Chevron test).
15
18 See Chevron, 467 U.S. at 842. One purpose of agency interpretations is to articulate
standards that are too technical or controversial for Congress to handle, not to replace a clearly
stated Congressional purpose. See id. at 843-44 (noting that the effectiveness of an
administrative agency "necessarily requires the formulation of policy and the making of rules
to fill any gap left, implicitly or explicitly, by Congress.").
19
See id.at 842. This approach can be dangerous, for it may allow a judge to circumvent
reasonable agency guidance by announcing a plain statutory meaning, even when such
meaning is much less than plain. See Thomas W. Merrill, Textualism and the Future of the
1999]
JUDICL4LDEFERENCETO THE EEOC
1537
Both of the above models relate to the varying amounts of deference that
courts will afford to administrative agencies. Therefore, earned and compelled
deference represent judicial reasoning in "deferential mode."2 0 A court in this
mode will recognize some inherent value of agency guidance, and seek a balance
of interpretive responsibility between itself and the administrative agency.2 1 This
is distinguished from the so-called "independent mode," in which the court gives
no special deference to agency guidance. 22 In the independent mode, a court
treats agency guidance no differently than other interpretive guidance such as
legislative history, or the litigants' arguments. 2 3 It is important to note that the
mode (whether deferential or independent) does not necessarily determine the
outcome in a given deference case. A court operating in independent mode is
free to follow an agency's view, 24 and a court in deferential mode may reject
it.25
Regardless of which mode (deferential or independent), or standard
(compelled or earned) might apply, judges who examine agency constructions
often face the seemingly endless question that riddled Chief Justice Marshall in
Marbury v. Madison:26 who has the ultimate duty to decide what the law is?
Marshall passed on to the judiciary the ultimate duty of constitutional
interpretation, 2 7 but the question remains: should the same interpretive duty
Chevron Doctrine,72 WASH. U. L.Q. 351,353-54 (1994) (pointing out the underlying tension
between the plain meaning approach and deference to agency interpretations, in that the former
may send judges into an "autonomous interpreter" mode that prompts excessive disregard for
agency guidance).
20
See Diver,supra note 9, at 559 (describing the "deferential mode").
21 See id. at 559 (describing the def&ential mode as a situation in which the court
"accord[s] a distinctive status to [the agency's] interpretation" ).
22
See id.at 552-59 (describing the "independent mode!).
23
See id. at 559 (describing the independent mode as a situation in which "the Court
specifies what the statute at issue means without regard for the way in which any other nonlegislative entity would interpret the statute"); Moot, supra note 12, at 215-16 (stating that,
under the independent mode, "the legislative will is changed only through amendment or
repeal of a statute and not through interpretation by judges or appointed administrators of
executive agencies").
24
Courts may cite an agency view in passing, and ultimately agree with it, without
according any special weight to that view. See cases cited infra note 87 and app. A (discussing
this phenomenon in the context of Supreme Court deference to EEOC interpretations).
25 A good illustration of a court rejecting agency guidance despite an overall deferential
posture is provided in Espinoza v. FarahManufacturing Co., 414 U.S. 86 (1975). There, the
Supreme Court stated that EEOC guidance was entitled to "great deference," but ultimately
rejected the guidance, recognizing that "deference must have limits." Id at 94.
26 5 U.S. 137 (1803).
27
See id. at 177 ("It is emphatically, the province and duty of the judicial department, to
1538
OHIO STATE LA WJOURNAL
[Vol. 60:1533
apply to statutory interpretation when an agency has exercised its delegated
power to interpret the statute at issue? Therefore, the basic problem of separation
ofpowers lingers at the foundation of any deference debate 2 8
A more fundamental problem arising from agency guidance is whether the
United States Constitution allows any delegation of legislative power to an
agency. The Constitution states that "All legislative Powers herein granted shall
be vested in a Congress .... 29 On its face, this provision seems to preclude any
delegation, and thus, any discussion about different models and modes of
deference would be moot. However, the constitutional provision known as the
Necessary and Proper Clause provides some justification for delegation of
legislative power to administrative agencies. The Clause states: "The Congress
shall have Power... to make all Laws which shall be necessary and proper for
carrying into Execution the foregoing Powers, and all other Powers vested by
this Constitution in the Government of the United States, or in any Department
or Officer thereof."30 The argument which follows is that delegation to agencies
that possess superior expertise over a matter at issue is "necessary and proper"
for the most efficient functioning of the legislature. 3 1 Yet, to satisfy this
constitutional requirement, Congress must delegate authority in a specific
fashion to ensure that the delegatee's legislative power is not overbroad. 3 2 As
say what the law is.").
28
See PETER L. STRAUSS et al., ADMNISTRATIVE LAw 51-52 (outlining the basic
constitutional problems arising from delegation and exercise of agency power). Professor
Martin Shapiro also provides an interesting theoretical perspective on the role of agencies in a
democratic system. See Martin Shapiro, On Predicatingthe Future of Administrative Law,
REGULATION, May-June 1982, at 18, 19-25 (discussing the development of administrative law
beginning during the New Deal and explaining how the problems of allowing non elected
officials to make binding policy since have been addressed). Shapiro presents two different
views on agency deference. Under the "technocratic" perspective, the agency deserves great
deference because of its expertise. However, from a purely democratic perspective, agencies
insulated from the political process (and thus, the popular will) are threatening to democratic
ideals. See id. at 21-22 (explaining that insulated agencies are more likely to cater to those
organizations that are best able to "obtain [] and exploit [ access to government and to the
public at large.').
29
U.S. CONST. art. I, § I.
30Id. at § 8.
31 An example in which delegation is necessary is in the environmental context. Congress
can make a decision to favor social over industry interests, but could not possibly determine
precise amounts of chemicals that should be allowed into the atmosphere, water, etc. See
generally Chevron, U.S.A., Inc. v. Natural Resources Defense Council, 467 U.S. 837 (1984).
32 Under the guise of the "delegation doctrine," the Supreme Court has traditionally
struck down, as unconstitutional, statutes that make sweeping delegations of legislative
authority to the President or an administrative body. See, e.g., A.L.A. Schechter Poultry Corp.
v. United States, 295 U.S. 495, 541-42 (1935); (striking down section 3 of the National
JUDICIALDEFERENCETO THE EEOC
1539
long as Congress upholds its end of the constitutional bargain, courts will most
likely recognize delegation of legislative power to agencies. Beyond such
recognition lies the question of deference. How do courts respond to agency
action after proper delegation? In particular, do courts respond to EEOC action
pursuant to proper delegation as favorably as other administrative agencies?
A. EarnedDeference: The Skidmore v. Swift Standard
The earned deference standard applies when delegation of rulemaking
authority to an agency is not clear from the empowering statute33 or when
delegation exists but the agency does not exercise its delegated authority to
promulgate binding rules.34 If an agency's action falls into either of these
categories, the agency must earn judicial deference by satisfying certain
judicially-crafted standards.35 In an earned deference case, a court may look to
Industrial Recovery Act of 1933 as an unlawful delegation of legislative power to the
President). Justice Rhenquist, concurring in Industrial Union Department, AFL-CIO v.
American PetroleumInstitute, noted that the delegation doctrine serves three basic functions.
Id. at 685. It ensures that (1) the most important policy choices are made by Congress; (2)
Congress will provide the recipient of authority with an "intelligible principle" to guide the
exercise of its delegated power, and (3) the courts will be able to exercise judicial review over
Congress's delegation of authority to determine whether it comports with basic constitutional
principles. See id. at 685-86. Much ink has been spilled by scholars in debate over the
propriety of the delegation doctrine. CompareJOHN HART ELY, DEMOCRACY AND DISTRUST, A
THEORY OF JUDICIAL REvImv
131-34 (1980) (arguing that delegation of "hard decisions" to
agencies through vaguely worded statutes is "undemocratic' because of the lack of
accountability for agency action); with Jerry L. Mashaw, Prodelegation: Why Administrators
ShouldMake PoliticalDecisions, 1J.L. ECON. & ORG. 81, 95-99 (1985) (arguing that political
accountability exists in the administrative context through presidential elections, and that
administrators operate within proper procedural and legal constraints).
33 For example, the EEOC did not receive complete rulemaking authority under Title VII,
thus, interpretations of that statutory provision fall under the earned deference standard. See
supra note 5 (discussing incomplete grant of regulatory authority to the EEOC under Title
VII).
34 Under the ADA, the EEOC did receive complete authority to promulgate binding rules.
See Bragdon v. Abbot, 118 S.Ct. 2196, 2209 (1998) (recognizing 42 U.S.C. § 12116 as the
source of ADA rulemaking authority). However, the EEOC may choose not to exercise its
authority when promulgating "interpretive" or "enforcement" guidance. Despite the name,
EEOC interpretive guidance may arguably transcend the earned deference category because it
was subject to APA notice and comment procedures. Therefore, enforcement guidance is a
better example of a situation in which the EEOC chooses not to exercise its rulemaking
authority, and thereby triggers the earned deference standard.
351Te reference to an agency "earning judicial deference reflects the reality that some
agency guidance is not inherently entitled to judicial deference. Such guidance may be
subjected to a reliability test. Depending on the result of that test, a judge in his or her
1540
OHIO STATE LAWJOURNAL
[Vol. 60:1533
factors set forth in Skidmore v. SWift 36 to determine the weight given to agency
37
guidance.
The interpretive question in Skidmore was whether time spent by employees
waiting on an employer's premises to respond to fire calls, in addition to regular
daytime work, constitutes "working time" under the Fair Labor Standards Act
(FLSA).3 8 Congress did not delegate rulemaking authority over the FLSA to any
agency,3 9 but it did empower an Administrator to explore working conditions in
industries subject to the Act, and to enjoin any unlawful employment practices 4 0
The Court determined that the Administrator's views are not controlling upon the
courts, but "do constitute a body of experience and informed judgment to which
41
courts and litigants may properly resort for guidance."
The Court then announced its famous test for determining the weight of non
controlling agency guidance. The weight of such an agency interpretation will
depend upon "the thoroughness evident in its consideration, the validity of its
reasoning, its consistency with earlier and later pronouncements, and all those
factors which give it power to persuade, if lacking power to control." 42 After
applying the above factors, the Court deferred to the Administrator's
interpretation, and reversed the district court's decision that waiting time is not
43
"working time" under the FLSA
Although numerous other factors not mentioned in Skidmore may affect the
weight of an interpretation, 4 the bottom line in an earned deference case is that a
discretion, may accept or reject the agency's view. Thus, some agency guidance must "earn"
deference. For an example of another commentator's usage of the phrase "earned" deference,
see Nicholas S. Zeppos, JudicialReview of Agency Action: The Problems of Commitment,
Non-Contractability,and the ProperIncentives,44 DuKEL.J 1133, 1153 (1995).
36 323 U.S. 134 (1944).
37
See id. at 140.
38 See id.at 135-37.
39
The Court pointed out that the absence of an explicit grant of rulemaldng authority
preserved interpretive power for the courts. "Congress did not utilize the services of an
administrative agency to... determine in the first instance whether particular cases fall within
or without the Act. Instead, it put this responsibility on the courts." Id. at 137.
40
See id. at 137-38 (discussing the Court's understanding of the role of the Office of
Administrator).
41 Id. at 140.
42
Id
43 See id.
44 Professor Colin S. Diver has documented other factors cited by the Supreme Court that
affect the weight of an agency interpretation. They include: (1) whether the agency
construction was contemporaneous with the passage of the statute; (2) whether the construction
is of longstanding application; (3) whether the public has relied upon the construction; (4)
whether the construction involves a matter of public controversy; (5) whether the agency has
1999]
JUDICIAL DEFERENCETO THE EEOC
court has great discretion in determining whether an agency has earned its
deference. If a court finds that a non binding agency rule suffers from some
procedural infirmity, or is not consistent with what the court believes is
Congress's intent, then the court may disregard the agency interpretation and
45
substitute its own views as to what the statute means.
B. CompelledDeference in the Chevron Era
In Chevron US.A. Inc. v. Natural Resources Defense Council,46 a
unanimous Supreme Court articulated what has become known as the compelled
deference standard. 47 The issue in Chevron was whether the court of appeals was
correct to set aside, as "contrary to law," regulations promulgated by the
Environmental Protection Agency (EPA) that defined the term "stationary
source" from the Clean Air Act.4 8 After articulating and applying its notorious
deference test, the Court deferred to the EPA's "reasonable policy choice," and
reversed the court of appeals' decision. 4 9
Justice Stevens articulated a two-step test for determining when and how to
defer to agency regulations. The first step is simple: if the plain language of the
statute is clear, then a court is obliged to follow that language without deference
to agency regulations.5 0 Under the second step, if the statute is ambiguous, then
rulemaking authority; (6) whether the interpretation is based on expertise or involves a
technical or complex subject; (7) whether agency action is necessary to set the statute in
motion; (8)whether Congress was aware of the agency interpretation and failed to repudiate it;
and (9) whether the agency has expressly addressed the application of the statute to its
proposed action. For a complete list of factors, including case citations, see Diver, supranote 9,
at 562 n.95.
45
See, e.g., Public Employees Retirement Sys. of Ohio v. Betts, 492 U.S. 158, 175
(1989) (rejecting the EEOC interpretation of the word "subterfuge' in the Age Discrimination
in Employment Act, and substituting the previous Supreme Court interpretation).
46 467 U.S. 837 (1984).
47
"Compelled deference" is used to describe the type of deference given to agency rules
that carry the force of law and satisfy the reasonableness requirement of the Chevron test. See
infra notes 50-52 and accompanying text. Chevron itself does not use the phrase "compelled
deference," rather, it has been used by several commentators discussing Chevron deference.
See e.g., Jerome Nelson, The Chevron Deference Rule and JudicialReview ofFERC Orders, 9
ENERGYL.J. 59,71 (1988); Yavelburg, supranote 11, at 186.
48
See Chevron, 467 U.S. at 842 (stating that the Court "must determine whether the
Court of Appeals' legal error resulted in an erroneous judgement on the validity of the
regulations").
49
See id. at 866 (holding agency's interpretation is a "permissible construction of the
statute," which should be accepted by courts, "who unlike the political branches, have no
constituency" ).
50
See id. at 842-43.
1542
OHIO STATE LA WJOURNrAL
[Vol 6:1533
the "question for the court is whether the agency's answer is based on a
permissible construction of the statute." 5 1
With regard to the second step, Justice Stevens further provides:
If Congress has explicitly left a gap for the agency to fill, there is an express
delegation of authority to the agency to elucidate a specific provision of the
statute by regulation. Such legislative regulations are given controlling weight
52
unless they are arbitrary, capricious, or manifestly contrary to the statute.
Under the above test, a court is compelled to follow "legislative regulations" for
which an explicit grant of authority exists, except in the rare cases in which the
53
regulation is deemed arbitrary, capricious, or manifestly contrary to the statute.
Little discretion remains for the reviewing court under this standard.
One notable problem arising from Chevron is the fact that its deference
standard does not appear to cover non binding or "interpretive" rules; it only
applies to "legislative regulations." 54 This is problematic because "legislative
regulations" comprise only a portion of an agency's rulemaking efforts. 55 If
Chevron is read to apply only to binding regulations, then its impact on judicial
deference is actually quite narrow, because some of the most contentious
deference issues arise from non binding agency guidance. The majority of the
Supreme Court has nonetheless been content to exclude interpretive rules from
56
Chevron's scope, but support for this view is not unanimous.
51 Id. at 842-43.
52
Id.at 843-44. Chevronwas not the first Supreme Court case to announce the deference
standard for legislative regulations. See, e.g., United States v. Morton, 476 U.S. 822, 834
(1984); Schweiker v. Gray Panthers, 453 U.S. 34, 44 (1981); Batterton v. Francis, 432 U.S.
416,424-26 (1977); American Tel. & Tel. Co. v. United States, 299 U.S. 232,235-37 (1936).
53
Professor Merrill, in an empirical study, found very few cases in which the Supreme
Court struck down regulations on the ground that they were adopted in an arbitrary and
capricious manner as opposed to being in conflict with the statute. See Merrill, supranote 12,
at981 n.51 (1992).
54
Interpretive rules are rules for which no congressional authority exists or rules
promulgated without a proper exercise of delegated authority. See Robert A. Anthony,
Interpretive Rules, Policy Statements, Guidances, Manuals, and the Like-Should Federal
Agencies Use Them to Bind the Public, 41 DUKE L.L 1311, 1313-14 (1992) (flaming the
distinction between legislative and interpretive rules). The two step deference test announced in
Chevron only applies to "legislative regulations." Chevron, 467 U.S. at 844.
55 Inaddition to legislative regulations, an agency may issue interpretive guidance or
policy statements. See DAVIs & PIERCE, supranote 7, at 228-250 (distinguishing various rulemaking efforts).
56
See EEOC v. Arabian Am. Oil Co., 499 U.S. 244,260 (1991) (Scalia, J., concurring)
(arguing that even non controlling agency guidance should be entitled to deference if
reasonable). Some lower federal courts have interpreted Chevron's second step to apply to
1999]
JUDICILDEFERENCETO THEEEOC
1543
C. Legislative Versus InterpretiveRules
57
Legislative regulations are grounded in a congressional grant of authority,
and enacted pursuant to the Administrative Procedure Act (APA)58 while
interpretive rules are exempt from these requirements. 59 The purpose of this
exemption is to "allow agencies to explain ambiguous terms in legislative
enactments without having to undertake cumbersome proceedings." 60 When
courts face difficulty distinguishing between the two, they generally focus upon
interpretive rules. See American Fed'n of Gov't Employees v. Fed. Labor Relations Auth., 778
F.2d 850, 861 (D.C. Cir. 1985) (accepting Federal Labor Relations Act (FLRA) "Interpretation
and Guidance" and holding that the agency may disapprove of contracts containing provisions
ordered by Federal Service Impasses Panel-which is contrary to law); Conley v. Brewer, 652
F. Supp. 106, 109-10 (W.D. Wis. 1986) (accepting Parole Commission's interpretive
regulation and explanatory statement that "good time" credits expire when prisoners are
released on parole); cf.Robert A. Anthony, Which Agency InterpretationsShould Bind Citizens
and the Courts?, 7 YALE J. ON REG. 1, 56-57 (1990) (suggesting that interpretive rules should
be entitled to stronger deference if they are subjected to notice and comment).
57 The ADA, Title I, provides an example of a clear grant of rulemaking authority to the
EEOC. "Not later than 1 year after July 26, 1990, the Cormmission shall issue regulations in an
accessible format to carry out this subchapter ....
"42 U.S.C. § 12116 (1994).
58 The APA sets forth the procedural requirements necessary for promulgation of a
binding agency rule. See 5 U.S.C. § 553 (1994) (setting forth publicity requirements for
agencies promulgating binding regulation). The APA distinguishes between formal and
informal rulemaking, either of which could satisfy the compelled deference standard if a
congressional grant ofrulemaking authority is explicit Compare 5 U.S.C. § 553, with 5 U.S.C.
§ 556 (1994). Informal rulemaking (also called "notice and comment rulemaking") is subject to
section 553 of the APA which requires an agency to afford "interested persons an opportunity
to participate in the rule making through submission of written data, views, or arguments with
or without opportunity for oral presentation." 5 U.S.C. § 553. This notice and comment
procedure takes place through publication of an initial draft of the regulations in the Federal
Register. See 5 U.S.C. § 553(b). On the other hand, formal rulemaking (which is less common
than the informal variety) is governed by sections 556 and 557, which require certain rules to
be made "on the record" after opportunity for an agency hearing. 5 U.S.C. §§ 556,557 (1994).
59
The APA states that notice and comment requirements do not apply to "interpretative
(sic] rules, general statements of policy, or rules of agency organization, procedure, or
practice." 5 U.S.C. § 553(bXA). One scholar suggests that interpretive rules should be subject
to the notice and comment process. See Anthony, supra note 54, at 1376 ("It would champion
the worthy precepts of the APA, however, if in certain circumstances agencies would
voluntarily make use of notice-and-comment rulemaking procedures to develop interpretive
rules."). The EEOC, when promulgating the "Interpretive Guidance" for its ADA regulations,
did exactly what Professor Anthony suggests: it subjected that guidance to notice and
comment. However, as discussed later in this Note, this tactic was not entirely successful in
securing deference. See infra Part IV.C.I.
60
American Hosp. Ass'n v. Bowen, 834 F.2d 1037, 1045 (D.C. Cir. 1987).
1544
OHIO STATE LAWJOURNAL
[Vol. 60:1533
the rule's impact on interested parties, rather than its form or origin.6 1
A rule's origin could be determined by its procedural history,62 or by a more
obvious indicator--4ts name.63 To determine impact, a court may consider
whether the rule creates new rights, produces significant effects on private
interests, or imposes obligations on the public. 64 Only a legislative rule
supported by a congressional grant of authority (either explicit or implicit) is
presumed to have such powerful effects. An interpretive rule, on the other hand,
"simply states what the administrative agency thinks the statute means, and only
'reminds affected parties of existing duties.' 65 Therefore, if an agency rule
produces powerful and binding effects, it may be properly considered
66
"legislative," even if the origin of the rule suggests otherwise.
61
This is known as the "legal effect" test. See JEFFREY S. LUBBERS, A GUIDETO FEDERAL
61-64 (3d ed. 1998). Lubbers also outlines another test, the "binding
nom" test, which, as he notes, is better suited to distinguishing legislative-type rules from
policy guidance. See id. at 64.
62 'MeAPA provides that legislative rules must be subjected to the notice and comment
process. See 5 U.S.C. § 553 ("General notice of proposed rulemaking shall be published in the
Federal Register... !).
63 The most obvious way for an agency to show that it is not exercising its delegated
rulemaking authority is to label a rule as 'interpretive." The EEOC made the choice (or
arguably, the mistake) of calling its ADA regulatory appendix "Interpretive Guidance.'
However, this choice may not be fatal, for as one court noted: "[Tjhe agency's own label, while
relevant, is not dispositive." General Motors Corp. v. Ruckelshaus, 742 F.2d 1561, 1565 (D.C.
Cir. 1984).
64
See Chrysler Corp. v. Brown, 441 U.S. 281, 302-03 (1979) ('We described a
substantive rule-or a 'legislative-type rule,'--as one 'affecting individual rights and
obligations.' ") (citations omitted) (quoting Morton v. Ruiz, 415 U.S. 199, 232, 236 (1974);
General Motors Corp. v. Ruckelshaus, 742 F.2d at 1565 ("[1]f by its action the agency intends
to create new law, rights or duties, the rule is properly considered to be a legislative rule:). For
similar pronouncements, see Flagstaff Med. Ctr. v. Sullivan, 773 F. Supp. 1325, 1351-52 (D.
Ariz. 1991), aff'd in part, rev'd in part, 962 F.2d 879 (9th Cir. 1992); Klingler v. Yamaha
Motor Corp., 738 F. Supp. 898, 903 (E.D. Pa. 1990).
65
Ruckelshaus, 742 F.2d at 1565-66 (citing Citizens to Save Spencer County v. United
States Envtl. Protection Agency, 600 F.2d 844, 876 & n.153 (D.C. Cir. 1979)).
66 Judge Kenneth Starr focused on impact in drawing the distinction between legislative
and interpretive rules, but recognizes the complexities of this standard. He declared that a
"legislative rule is recognizable by virtue of its binding effect." Alaska v. Dept. of Trans., 868
F.2d 441, 445 (D.C. Cir. 1989). He further noted that this "definitional principle... is hardly
self-executing," and cited numerous "factors" which may be necessary to properly categorize a
rule. Id. at 466. The factors that he used to determine whether the agency rule had binding
effect included: mandatory language, publication in the Code of Federal Regulations,
limitations upon agency discretion, and the ability of the agency to prosecute persons for
noncompliance with agency guidelines. See idat 446-47.
AGENCY RULEMAKiNG
1999]
JUDICIALDEFERENCETO THEEEOC
1545
However academic the distinction between legislative and interpretive rules
may seem, it may be critical in a deference case because legislative rules that
have been subjected to notice and comment procedures, qualify for Chevron
deference while interpretive rules usually do not. 67 This distinction also raised
the question of whether rules that are clearly interpretive in nature should
nonetheless qualify for Chevron deference if they are subjected to the notice and
comment process. This question will arise again later in a discussion of EEOC
"Interpretive Guidance." 6 8 For now, the focus remains on theoretical models of
deference, and more specifically, the effects of Chevron.
D. The PracticalEffects ofChevron on Agency Deference
Chevron was cited in approximately 3,500 federal cases from 1984 to
1996. 69Although the Chevron standard has been proclaimed as a "revolution on
71
paper,"'70 its practical effect on deference decisionmaldng is unclear.
According to an empirical study by Professor Thomas Merrill, the rate of
Supreme Court deference to agency interpretations did not change significantly
67
No language in Chevron explicitly precludes its application to interpretive rules, but the
majority of the Court has "read in" that preclusive effect. See supra notes 54-56 and
accompanying text (discussing preclusive effect of Chevron).
68 This Guidance, which was published as an appendix to the ADA regulations, was
subjected to notice and comment requirements. See infra Part IV.C.1.
69 A LEXIS search conducted on December 5, 1996 in the GENFEDS:COURTS
database, which includes all published federal decisions, generated 3,567 cases citing Chevron.
See David M. Gossett, Chevron, Take Two: Deference to Revised Agency Interpretationsof
Statutes, 64 U. Ci. L. REV. 681, 695 &n.67 (discussing the quantity of Chevron citations).
70 The Chevron test is characterized by Professor Merrill in this manner because, prior to
1984, the Supreme Court had no consistent standard for determining when to defer to agency
interpretations. See Merrill, supra note 12, at 971-72. During this period, the approach was
"pragmatic and contextual," and the results, "could range from 'great' to 'some' to 'little"'
deference. Id. at 972. Chevron presents a clear, categorical test for deference, hence, it is aptly
described as a "revolution on paper." For another perspective on how Chevron strengthened
the deference principle, see generally Kenneth W. Starr, JudicialReview in the Post-Chevron
Era, 3 YALE J. ON REG. 283 (1986). Judge Starr discusses three ways in which Chevron
buttressed the deference principle: first, it removed the ambiguity arising from a long line of
cases, some of which called for deference and others which disregarded it altogether, second, it
curbed judicial discretion to invalidate agency guidance based on perceived inconsistencies
with congressional policies; and third, it specified certain conditions under which courts are
compelled to give controlling weight to agency interpretations. See id. at 292-99.
71 Upon examination of post-Chevron trends, Professor Merrill concluded that "[T]he
overall picture suggests that the judicial understanding that informs the deference question is
probably more confused today than it has ever been." Merrill, supranote 12, at 980.
1546
OHIO STATE LAWJOURNAL
[Vol. 60:1533
after Chevron.72
Merrill's suggestion that Chevron's practical effects are negligible contrasts
sharply with another empirical study by Professors Peter Schuck and Donald
Elliot. The latter study surveyed the effects of Chevron on lower courts and
found that deference rates increased sharply after Chevron.73 However, as
Merrill points out, this study does not necessarily conflict with his data because
"[1]ower courts probably take Supreme Court opinions more seriously than does
the Court itself."7 4 Merrill further asserts that the effect of Chevron in lower
75
courts may have been only temporary.
Despite its interpretive problems, Chevron does provide valuable
background for the following distinction between legislative and interpretive
rules.
III. AN EMPIRICAL PERSPECTIVE ON JuDICIAL DEFERENCE TO THE EEOC
Having provided some background on the different models of judicial
deference, I now consider how those models have been applied to EEOC
guidance. I first present empirical data collected by Professor Thomas Merrill
regarding Supreme Court deference to all agencies. This will serve as a baseline
against which I compare the rate of Supreme Court deference to the EEOC. If
EEOC deference rates are significantly lower than the baseline figure, then one
could claim (albeit not with absolute certainty) that the EEOC receives less than
its fair share of deference. I caution the reader, as many scholars have cautioned
me, that empirical research in the area of agency deference is inherently
72
Professor Merrill compared Supreme Court deference cases in two time periods: 1981
to 1983 (pre-Chevron), and 1984 to 1990 (post-Chevron). In the first block, he found that the
agency view was "accepted" in 75% of 45 cases. In the second block the rate was 70% of 90
cases. Although these statistics may be misleading for multiple reasons (the most obvious of
which is small sample size), one could safely say that this study, if accurate, softens the flame
of the "Chevron Revolution." See id. at 980-83.
73 The study reported strong evidence that lower court decisions granting deference were
affirmed at a rate of 71% in the six months prior to Chevron, compared to an 81% affirmance
rate during a six-month period in 1985, after Chevron. See Peter H. Schuck & Donald Elliott,
To the Chevron Station: An EmpiricalStudy ofFederalAdministrativeLaw, 1990 DuKE LJ.
984, 1029-32 (1990). However, the article also reports only a 76% deference rate in 1988
(with remands twice as frequent as reversals, whereas in 1984-85 they had been roughly
equal). By that year, according to Shuck and Elliot, "the Supreme Court had appeared to recede
from the position of extreme deference," and the lower courts apparently followed suit. Id at
1037.
74
Merrill, supranote 12, at 984.
75
See id.at n.62 (citing instances in which Shuck and Elliot recognize the weaknesses of
their own conclusions and offer reasons why the post-Chevron change might not endure).
1999]
JUDICIAL DEFERENCETO THEEEOC
1547
unreliable. In response to this problem, I have attempted to reinforce my
empirical research with a more narrow deference inquiry. I examine some of the
leading deference cases in the context of the Civil Rights Act, the ADEA, and
the ADA-with a special focus on the ADA.
A. OverallAgency DeferenceAfter Chevron: The Baseline
Professor Thomas Merrill examined Supreme Court deference to
administrative agency interpretations in a "pre" and "posf' Chevron
framework.7 6 Between 1981-1983 (3 years before Chevron was decided), the
"total cases involving [a] deference question" was 45,77 and the "agency view
[was] accepted" in 34 of those cases, 78 yielding a 75% deference rate.7 9 Between
1984-1990 (post-Chevron), 63 out of 90 cases involving deference, or 70%,
resulted in acceptance of the agency view.80 Thus, the average deference rate
from these two time periods is approximately 72%.
Before drawing any conclusions from these numbers, I must point out what I
believe are the major weaknesses of the above survey. First, the data does not
distinguish between different agencies, which is critical because different
agencies might receive different treatment in the courts. For example, one cannot
necessarily assume that a 72% acceptance rate would apply equally to the SEC
81
and the EEOC.
76
See generally Merrill, supranote 12 (comparing Supreme Court deference to agency
interpretation in the three terms prior to Chevron with Court deference in the seven terms after
Chevron).
77
See id. at 982 tbl.2. This title category, although simple on its face, presents certain
methodological problems recognized by Merrill in an elaborate footnote. See id. at 981 n.51.
The first of those problems is that many cases exist where the Court substitutes its own
judgment for that of the agency and fails to even mention deference. However, it would be
difficult to determine all the cases in which the Court could have deferred. Thus, Merrill
excluded these so-called "potential deference" cases from his sample. The second problem was
caused by cases which "present slight variations on the question of deference," such as agency
interpretations of its own regulations or treaties; or challenges to agency policies on the ground
that they were adopted in an arbitrary and capricious manner (as opposed to being in conflict
with the statute). Merrill excluded these cases, which are relatively few in number, to
"minimize complexity." See id.
78 See id at 982 tbl.2. Again, this title category requires some explanation. Merrill points
out that "accepting the agency interpretation is not the same as ruling in favor of the
govenment." Id. at 981 n.52. He goes on to cite Bob Jones University v. United States, 461
U.S. 574 (1983), a case in which the Solicitor General disclaimed the agency view, but the
Court deferred to it.
79
See Merrill, supranote 12, at 982 tbl.2.
8
0 See Id at 981 tbl.1.
81 From one broad empirical study, one could infer that the SEC and EEOC receive
1548
OHIO STATELA WJOURNAL
[Vol. 60:1533
Second, the surveys do not distinguish between different types of agency
guidance. Agencies could issue regulations, interpretive guidance, interim
enforcement guidance, or technical assistance. 82 A regulation promulgated under
express authority from Congress will almost always receive deference unless
arbitrary or capricious, while interim enforcement guidance is easily
disregarded. 83 Thus, deference rates will vary widely depending on the type of
agency guidance at issue.
Another
fundamental
problem
arises
in
determining
the
category--'deference" or "no deference"--into which the agency guidance is
placed. Often the Court will not explicitly pay deference, but merely cite some
agency guidance in passing or with mild approval. Placing these cases in the
"deference" category may ultimately shift the balance of cases in a misleading
way. However, creating a standard to distinguish them would be a technical
nightmare.
Despite these problems, empirical research brings us closest to the truth
about agency deference. The "leading case" approach, which involves analysis of
a discrete sample of representative cases, is another method often used to identify
trends in Supreme Court jurisprudence. 84 This approach, although certainly more
convenient for the researcher, is much less reliable than empirical research for
similar rates ofjudicial deference. See generally Reginald S. Sheehan, AdministrativeAgencies
and the Court: A Reexamination of the Impact ofAgency Type on Decisional Outcome, 43 W.
POL. Q. 875, (1990) (finding similar deference rates between "economic" and "social"
agencies). Professor Sheehan examined over 800 Supreme Court cases from 1953-1988 and
placed each case into one of two categories: economic agencies such as the FCC, NRC, and
SEC; and social agencies such as the EPA, EEOC, FDA, NLRB, and OSHA. See id at 87778. He further broke them down according to the Court period: Burger and Warren Court.
After further categorization into "liberal" or "conservative" agency action, he found that,
despite significant vacillation in deference rates between different Courts, the average
deference rate for economic and social regulations was roughly the same. See id. at 879-81.
The rate for social regulations was approximately 75% compared to a 73% rate for economic
regulations. For a detailed chart of these findings, which properly separates the data according
to the time period and the type of guidance, see STRAUSS, supranote 28, at 514.
82 These types are listed in order of strength, with regulations being most persuasive and
technical assistance being the least persuasive. See CCR REPORT ON ADA TITLE I, supranote
1, at 66-77 (discussing different types of agency guidance and their persuasive values).
Although the empirical studies at issue do discard regulations reviewed under the arbitrary and
capricious standard, see Merrill supra note 12, at 981 n.51, they still do not distinguish
between other types of agency guidance.
83
Interim enforcement guidance is entitled to very little deference. This type of guidance
is discussed below in the context of the ADA. See infra Part IV.C.2.c.
84
See Shuck & Elliot, supra note 73, at 1060 (contrasting the "empirical" and "leading
case' approaches).
1999]
JUDICIALDEFERENCETO THEEEOC
1549
determining legal trends.85 This Note takes a "belt-and-suspenders" approach by
first, presenting empirical data regarding deference to the EEOC; and second,
conducting an analysis of a discrete sample of "leading" EEOC deference cases.
B. Supreme CourtDeference to the EEOC: Beyond the Baseline
One purpose of this Note is to determine whether the EEOC receives less
deference than other administrative agencies. After examining the total number
of Supreme Court cases involving deference to EEOC guidance and comparing
that data to the baseline figure of seventy-two percent, this Note concludes that
the EEOC receives a considerably low level of deference.
Between the years 1964-1998, the Supreme Court delivered 28 cases
involving deference to EEOC guidance.86 The Court accepted the EEOC view in
85
According to Shuck and Elliot, who conducted an empirical survey of lower court
deference cases, the leading case approach "is quite unsatisfactory" because "[i]t posits that the
'law' .. . can be captured in a few supposedly exemplary cases." Id. at 1060 (citation omitted).
The leading case approach may be the only altemative when dealing with a discrete issue and a
small sample of cases. Considering the large number of Supreme Court deference cases and the
breadth of issues addressed in those cases, the leading case approach is inherently unreliable in
determining deference trends.
86 These cases were retrieved from WESTLAW (and double-checked with a LEXIS
search). The search terms included "EEOC," "deference," "defer," "administrative,"
"regulation," "guidance," "guidelines," and various others. In an attempt to mirror Professor
Merrill's methodology (which serves as the baseline for this data) I discarded numerous cases
before arriving at a sample of 24. See Merrill, supra note 12 at 981 n.51 (discussing Merrill's
system of categorizing agency deference cases). The following cases were discarded: (1) cases
involving the validity of agency arguments advanced by the agency during adjudications; (2)
"potential deference" cases, i.e., those in which the Court substitutes its own judgment for that
of the agency without citing the agency's guidance or discussing the possibility of deference;
(3) cases where agency guidance is challenged on the ground that it is adopted in an arbitrary
and capricious manner, and (4) cases involving purely procedural rules adopted by the agency.
Furthermore, in keeping with Professor Merrill's methodology, I adopted a selection principle
that included all cases in which at least one Justice mentioned the possibility of deferring to an
agency interpretation. Therefore, if the majority substitutes its own interpretation without
mentioning the EEOC view, while a dissenting or concurring Justice points out that some
EEOC guidance was relevant and deserving of deference, I placed these in the "not accepted"
category. This occurred in the following cases: International Broth. of Teamsters v. United
States, 431 U.S. 324, 390-91 (1977) (Marshall, J., dissenting) ("While the Court may have
retreated from its prior view that interpretations are 'entitled to great deference,' I have not.");
Washington v. Davis, 426 U.S. 229, 263 (1976) (Brennan, J., dissenting) ("[T]he construction
of a statute by the agency charged with its administration is entitled to great deference.");
Phillips v. Martin Marietta Corp., 400 U.S. 542, 545-46 (1971) (Marshall, J., concurring)
("Tjhe Equal Employment Opportunity Conmission['s] ... regulations are entitled to great
deference."). Likewise, if the majority did not rely upon EEOC guidance in its determination,
but a concurring Justice pointed out that EEOC guidance would support the majority decision,
1550
OHIOSTATE LAWJOURNAL
[VoL 60:1533
15 of those cases; 87 and rejected the EEOC view in 13 cases. 88 This yields a
deference rate of approximately 54%. Compared to the baseline figure of 72%,
this number is considerably low.89 According to these statistics, one could fairly
characterize the EEOC as a "second class agency," at least in the eyes of the
Supreme Court.
The most obvious problem with the above sample is its small size.9 0
However, it is important to note that the above survey targets a discrete sample
of cases, i.e., those in which the court is judging the validity of published EEOC
I then placed such a case in the "accepted" category because the Courts holding was ultimately
consistent with the agency view. This occurred in only one case in my sample, Johnson v.
Transp. Agency, 480 U.S. 616, 642-47 (1987) (Stevens, J., concurring) (citing McDonald v.
Santa Fe Trail Transp. Co., 427 U.S. 273, 279-80 (1976)).
87 For a complete list of cases in the "accepted" category, see infra app. A. I adopted a
broad standard for determining whether the EEOC's view was "accepted." I included cases in
which the court simply cited agency guidance in passing, without offering any special
deference to that guidance. See, e.g., County of Washington v. Gunther, 452 U.S. 161, 190
(1981) (Rhenquist, J., dissenting) ("[W]hat little legislative history there is on the subject
...and the contemporaneous interpretation of the EEOC... indicates that Congress intended
to incorporate the substantive standards of the Equal Pay Act into Title VII.... ').In such
cases, the Court was operating in independent, not deferential mode, that is, it treated agency
guidance no differently than other extrinsic aids such as legislative history. For a brief
discussion of these two "modes," see Diver supra note 9, at 552-67. Nonetheless, even when
operating in independent mode, the Court often reached the same decision as the agency. I
would have preferred to exclude these cases altogether, because they do not represent the court
in a deferential posture toward the EEOC. However, because Merrill separated cases according
to whether the agency view was accepted or not accepted, and because his survey provides a
baseline for my EEOC data, I included cases in the "acceptance" category as long as the
agency view was consistent with the ultimate holding of the Court.
88 For a complete list of cases in the "rejected" category, see app.B. A common reason for
rejection of EEOC guidance, occurring in 5 out of 12 cases, is that the agency guidance
conflicts with the plain language of its empowering statute. See app. B, Cases 3, 4, 6, 9, 11.
The frequent use of this method of rejection supports Professor Merrill's theory that a textualist
approach is excessively employed to discredit agency guidance. See supranote 17 (discussing
Merrill's theory regarding the effect of textualism on agency deference). Another common
justification for rejection of EEOC interpretations of Title VII was the fact that the EEOC has
limited rulemaldng power under Title VII. See app.B, Cases 1, 2,4,9.
891 specifically avoided the phrase "significantly low" because I did not wish to connote
the more scientific notion of "statistical significance." I do not assert that the difference
between a 54% and 72% deference rate conclusively establishes the EEOC's "second class"
status. I only wish to point out that a 54% deference rate is suspect and thus, deserves further
attention.
90 No specific guidelines exist to determine the proper sample size for empirical legal
research. From a sample of 25 cases, one cannot draw secure conclusions about legal trends.
However, when examining a somewhat discrete class of Supreme Court cases, one cannot
expect great quantity.
1999]
JUDICIALDEFERENCETO THE EEOC
guidance (as opposed to the EEOC's adjudicatory positions). Thus, one cannot
expect this sample to be large, especially since the EEOC is a relatively young
agency, and the use of rulemairing as the primary means of agency expression
did not fully develop until the late 1970s and early 1980s. 9 1
Despite the fact that the above statistics may be incomplete, they do provide
some perspective on the level of deference paid to the EEOC in comparison to
overall agency deference. To supplement this empirical data, the remainder of
this Note presents a more narrow inquiry of EEOC deference by focusing on the
treatment of some discrete and controversial EEOC provisions.
IV. DEFERENCE TO EEOC INTERPRETATIONS OF TITLE VII, THE ADEA,
ANDTHEADA
For each of the statutes it administers, the EEOC has a specific grant of
authority from Congress that defines its enforcement role. For each statute in the
following sections, I begin by tracing the EEOC's level of authority, and proceed
to discuss some of the most contentious deference issues arising from the statute
and its administrative interpretations. I devote the bulk of the following analysis
to the ADA, for it is the youngest statute from which many deference issues
arise.
A. Deference to EEOCInterpretationsof Title Vii
The EEOC was bom out of the Civil Rights Act of 1964.92 Because of great
tension surrounding the passage of the Act,9 3 the EEOC's power was limited to
"procedural regulations," not regulations with the force of law.94 President
Carter attempted to expand EEOC authority over Title VII through an executive
order.95 However, as one scholar noted, the order was ineffective because "[t]he
91 See STRAUSS, supranote 28, at 426 ('IThe 1970s and early 1980s saw a dramatic shift
away from agencies' traditional preference for case-by-case development of policy [i.e.,
adjudication].").
92
See Pub. L. No. 88-352, Title VII § 705, 78 Stat 258 (1964) (codified as amended at
42 U.S.C. § 2000e4 (1994)) (providing for composition, duties, and powers of the EEOC).
93
The Senate debate over the Civil Rights Act has been characterized as the "Longest
Debate." See WILLIAMN. ESKR DE, JR. & PHILIP P. FRICKEY, LEGISLATION 17 (2d. ed. 1995).
94
"The Commission shall have authority from time to time to issue, amend, or rescind
suitable procedural regulations to carry out the provisions of this subchapter." 42 U.S.C.
§ 2000e-12(a) (1994). Perhaps in an effort to compensate for the EEOC's lack of rulemaking
authority, Congress provides an exemption from liability under Title VII for a person acting "in
reliance on any written interpretation or opinion of the [EEOC] .... " Id § 2000e-12(b).
95
Executive Order No. 12106, which was issued pursuant to Carter's Reorganization
Plan No. 1 of 1978, stated that the "Equal Opportunity Employment Commission, after
1552
OHIOSTATE LA WJOURNAL
[Vol. 60:1533
President ...does not have the power to increase the stature of an executive
agency's regulations; the delegation doctrine reserves that power for
Congress." 96 Therefore, under Title VII, the EEOC is authorized to promulgate
only interpretive or procedural guidelines and not regulations with the force of
97
law.
Despite the limits placed on EEOC rulemaking power, the Supreme Court
held early EEOC guidance in high regard. This period of "great deference,"
however, was short-lived.
1. The Era of "GreatDeference"
The Court generally treated the EEOC favorably in its early deference cases,
but the EEOC was off to a slow start in Phillipsv. Martin Marietta Corp.98 This
was the first case where the Court considered the issue of deference to EEOC
guidance. It involved a claim of sex discrimination, which turned on the meaning
of the Title VII exception for discrimination based on a "bona fide occupational
consultation with all affected departments and agencies, shall issue such rules, regulations,
orders, and instructions and request such information from the affected departments and
agencies as it deems necessary and appropriate to carry out this Order." Exec. Order No.
12,106, 3 C.F.R. 263 (1979), reprinted in 42 U.S.C. § 200e-4nt (1994) (transferring
administrative functions to the EEOC). This was the same plan that successfully transferred
authority over the ADEA from the Department of Labor to the EEOC. See Reorg. Plan No. 1
of 1978, supranote 4 (discussing transfer of ADEA authority to EEOC).
96 Colker, supra note 11, at 139. The idea that only Congress can grant rulemaking
authority to an agency has its origins in Article I, Section I of the United States Constitution.
"All legislative Powers herein granted shall be vested in a Congress...." U.S. CoNsT., art. I,
§ 1. The President, as the head of the Executive Branch, could transfer authority from one
agency to another. In that situation, authority is not being created or destroyed--a power which
lies only in congressional hands.
97 The EEOC has promulgated interpretive guidelines targeted toward a discrete number
of discrimination issues: see 29 C.F.R. § 1604 (1998) (sex); 29 C.F.R. § 1605 (1998) (religion);
29 C.F.R. § 1606 (1998) (national origin); 29 C.F.R. § 1607 (1998) (employee selection); and
29 C.F.R. § 1608 (1998) (affirmative action). Nearly all of these guidelines have proved
contentious, but those addressing national origin are currently the subject of heated debate.
Compare Lisa L. Behm, ProtectingLinguistic Minorities Under Title VII The Need for
JudicialDeference to the EEOC Guidelines on DiscriminationBecause ofNationalOrigin,81
MARQ. L. REV. 569, 571-72 (1998) (arguing that the resolution of the conflict over Englishonly rules "lies in judicial deference to EEOC guidance"); with David T. Wiley, Whose
Proof.: Deference to EEOC Guidelines on Disparate Impact Discrimination Analysis of
"English-only" Rules, 29 GA. L. REV. 539, 543 (1995) (arguing that, "contrary to EEOC
guidelines, the complaining employee should bear the burden of proving an English-only rule
has a disparate impact).
98 400 U.S. 542 (1971) (per curiam).
1999]
JUDICALDEFERENCETO THE EEOC
1553
qualification" (BFOQ).9 9 The majority ultimately attached a broad interpretation
to the BFOQ exception-contrary to the narrow interpretation of the EEOC. 10 0
In concurrence, Justice Marshall objected to the majority's broad interpretation
of the BFOQ exception and argued that the EEOC interpretation, which was
entitled to "great deference," should have been controlling. 0 1
In Griggs v. Duke PowerCo.,10 2 the majority did offer great deference to the
EEOC.10 3 The issue in Griggs was whether abilities targeted in employee
performance tests had to be "significantly related to successful job
performance." 104 The EEOC's position was that such tests must relate to
"important elements of work behavior. ' 105 The Court found language in the
statute and its legislative history which supported the EEOC view and stated that
the EEOC guidelines were entitled to great deference.1 06 Ultimately, the Court
agreed with the EEOC, 10 7 concluding that Title VII precludes the use of testing
99 42 U.S.C. § 2000e-2(e)(1) (1994).
10 0
See Guidelines on Discrimination Because of Sex, 29 C.F.R. § 1604.1 (1970). In
these guidelines, the EEOC stated that the BFOQ exception should be construed narrowly, and
cited numerous examples of situations that do not satisfy the exception. These situations
include: (1) refusal to hire women based on the notion that the turnover rate for women is
higher;, (2) the notion that women are less able to assemble complex equipment or act as
aggressive salespersons; (3) higher costs for providing separate facilities for both sexes. For a
more comprehensive discussion of the BFOQ exception in the context of the Phillips case, see
Moot, supranote 12, at 222-23 & nn.78-86.
101 See Phillips,400 U.S. at 545 (1971) (Marshall, J., concurring). The concurrence cited
Udall v. Tallman, 380 U.S. 1, 16 (1965), in support of its "great deference" position. In Udall,
the Court offered great deference to an interpretation by the Secretary of Interior. See Udall,
380 U.S. at 16 ("When faced with a problem of statutory construction, this Court shows great
deference to the interpretation given the statute by the officers or agency charged with its
administration.').
102 401 U.S. 424 (1971).
103
Id. at 433-34 ("The administrative interpretations of the Act by the enforcing agency
is entitled to great deference.").
104
See id. at 426. The employer required a high school education and the passing of a
standardized general intelligence test as a condition of employment, although neither standard
was significantly related to successful job performance.
105
See id. at 433 n.9 (citing and discussing EEOC Guidelines on Employee Testing
Procedures, 29 C.F.R. § 1607.4(c) (1970)).
106
See id.at 433-34.
107 This language is not surplusage. The fact that the Court offered great deference does
not mean it must ultimately agree with the guidelines. For example, in Espinoza v. FarahMfg.
Co., 414 U.S. 86 (1973), the Court articulated the great deference standard for EEOC guidance,
which prohibited discrimination "on the basis of his citizenship" against a lawfully immigrated
alien. Id. at 94. However, recognizing that "deference has its limits," and that finding that the
guidelines conflicted with Congress's intent, the Court ultimately rejected the EEOC view. Id.
1554
OHIO STATE LA WJOURNAL
[Vol. 60:1533
procedures which are not substantially job-related. 10 8
A series of cases after Griggs revealed the Court's willingness to share
interpretive responsibility over Title VII with the EEOC. 10 9 For example, in
Albemarle Paper Co. v. Moody 10 the Court once again offered great deference
to EEOC guidelines regarding employment tests. 111 It accorded such deference,
despite the fact that the guidelines were not promulgated pursuant to formal
procedures established by Congress, 112 because they constitute "[t]he
I 13
administrative interpretations of the Act by the enforcing agency."
2. The Later Cases: Departurefrom DeferentialMode
Beginning with GeneralElectric Co. v. Gilbert,114 the Supreme Court began
to express some hostility toward EEOC guidance. The issue in Gilbert was
whether discrimination on the basis of pregnancy constituted sex discrimination.
The Court rejected EEOC guidelines which provided that disabilities caused by
gender should be treated in the same manner as other temporary disabilities. 115
The Court in Gilbert reasoned that because Congress "did not confer upon
the EEOC the authority to promulgate rules or regulations pursuant to [Title
108 See Griggs, 401 U.S. at 436 ("Congress has forbidden ... giving these devices and
mechanisms controlling force unless they are demonstrably a reasonable measure of job
performance.").
109 See EEOC v. Commercial Office Prods. Co., 486 U.S. 107 (1988) (recognizing
workshare agreements between the EEOC and state agencies); EEOC v. Associated Dry Goods
Corp., 449 U.S. 590, 604 (1981) (deferring to EEOC regulations governing the disclosure of
investigative information); Trans World Airlines, Inc. v. Hardison, 432 U.S. 63, 66 (1977)
(deferring to EEOC guidelines regarding the "religious needs of... employees"); Albemarle
Paper Co. v. Moody, 422 U.S. 405,431 (1975) (finding that "EEOC guidelines ... constitute
'[t]he administrative interpretation of the Act by the enforcing agency,' and consequently they
are 'entitled to great deference") (citation omitted).
110 422 U.S. 405 (1975).
111 Seeia at431.
112
See supra note 58 (describing APA notice and comment requirements).
11 3 Albemarle Paper Co., 422 U.S. at 431. The Court, in essence, applies the earned
deference standard to the guidelines because it offered deference to an interpretive rule with
procedural infirmities.
114 429 U.S. 125 (1976), reh 'g denied,429 U.S. 1079 (1977).
115 See id. at 143. The EEOC guideline at issue in Gilbert provides in pertinent part
"Disabilities caused or contributed to by pregnancy, miscarriage, abortion, childbirth, and
recovery therefrom are, for all job-related purposes, temporary disabilities and should be
treated as such under any health or temporary disability insurance or sick leave plan available
in connection with employment...." 29 C.F.R. § 1604.10(b) (1975).
JUDICILDEFERENCETO THE EEOC
1999]
VII], 11
6
courts may accord 'less weight to such guidelines
1555
than to
administrative regulations which Congress has declared shall have the force of
law.' 117 The Court then applied the traditional Skidmore v. SWft factors1 18 to
determine the weight of the guidelines and found that the guideline in question
"does not fare well under [those] standards."1 19
The extension of the Skidmore doctrine to the EEOC guideline in Gilbert
may give the appearance of deference because the Court could did not substitute
its own judgment in an entirely independent manner. However, the Skidmore test
presents a situation of minimal deference. 120 The weight assigned to an agency
guideline under Skidmore is dependent upon, inter alia, its "thoroughness" or
"validity," 12 1 which leaves great discretion to the reviewing Court. The bottom
line in many cases in which Slddmore factors provide the appearance of
deference is that the Court is not sharing its interpretive power with the agency.
No such balance existed in Gilbert, nor in subsequent cases in which the Court
122
continued to manifest its distrust of EEOC interpretations.
Regardless of any criticism directed toward the Court's rejection of EEOC
guidance in Gilbert,123 one cannot overlook the fact that the EEOC's
116 Gilbert,429 U.S. at 141.
117 Id.
118 323 U.S. 134 (1944). The Gilbert Court articulated the Skidmore test as follows: "The
weight of such ajudgment in a particular case will depend upon the thoroughness evident in its
consideration, the validity of its reasoning, its consistency with earlier and later
pronouncements, and all those factors which give it power to persuade, if lacking power to
control." Gilbert,429 U.S. at 142 (citing Skidmore, 323 U.S. at 140).
119 Id. at 142. The Court found that the guidelines were not rendered contemporaneously
with the passage of Title VII, that they conflicted with earlier EEOC announcements, that
legislative history conflicted with the EEOC position, and that the EEOC's interpretation of
Title VII conflicted with the interpretation of the Wage and Hour Administrator who was
charged with enforcement of the Equal Pay Act. See id at 142-45.
120 Professor Diver points out that Skidmore deference could be nothing more than
"respectful or courteous regard" because the "pedigree" of the agency's argument adds nothing
to the persuasive force inherent in its reasoning. See Diver, supra note 9, at 565. Therefore,
even though Skdmore prescribes standards for determining weight, the agency's interpretation
is not inherently more persuasive than any other argument placed before the court.
121 See Skidmore, 323 U.S. at 140.
122
See, e.g., EEOC v. Arabian Am. Oil Co., 499 U.S. 244 (1991); Washington v. Davis,
426 U.S. 229 (1976); Espinoza v. Farah Mfg. Co., 414 U.S. 86 (1975); see also Yavelberg,
supra note 11 (discussing trend of decreased agency deference after EEOC v. Arabian
American Oil Co., 499 U.S. 244 (1991)).
123 See Gilbert, 429 U.S. at 157-58 (Brennan, J., dissenting) ("For me, the 1972
guideline represents a particularly conscientious and reasonable product of EEOC deliberations
and, therefore, [it] merits our 'great deference.' Certainly, I can find no basis for concluding
that the guideline is out of step with congressional intent.").
1556
OHIO STATE LA WJOURNAL
[V:ol. 60:1533
enforcement power over Title VII is inherently weak. Therefore, a lack of
consistent deference to the EEOC in the Title VII context may be justified. A
lack of deference to EEOC interpretations of the ADEA and ADA, however,
cannot so easily be justified, because those interpretations are grounded in
explicit congressional grants of authority. 124 As illustrated in the following
section, the EEOC has not received significantly better treatment under the
ADEA. 125
B. Deference to EEOCInterpretationsofthe ADEA.
The Age Discrimination in Employment Act (ADEA)126 was enacted in
1967 to prohibit age-based discrimination against older individuals in the terms
or conditions of employment. 127 To carry out these goals, Congress explicitly
128
delegated authority to the EEOC to promulgate rules with the force of law.
Enforcement power over the ADEA was originally vested in the Department of
129
Labor, but was subsequently transferred to the EEOC by President Carter.
Despite this transfer of authority, the EEOC's rulemaking power under the
ADEA remains strong.130 Therefore, any EEOC regulations regarding the
124
See supra notes 3 and 4 (discussing statutory grants of authority to EEOC under the
ADEA and ADA). Although the EEOC did receive rulemaldng authority under the ADEA and
ADA, the agency may still choose to promulgate interpretive rules, for which it does not
exercise its delegated authority. Such rules are properly entitled to less weight.
12 5
See infraPart IV.B (discussing deference to EEOC interpretations of ADEA).
12 6
Age Discrimination in Employment Act of 1967, Pub. L. No. 90-202, 81 Stat. 602
(codified as amended at 29 U.S.C. §§ 621-634 (1994 & Supp. I1 1997)).
127
See 29 U.S.C. § 623(a) (1994).
28
1 "mhe Equal Employment Opportunity Commission may issue such rules and
regulations as it may consider necessary or appropriate for carrying out this chapter, and may
establish such reasonable exemptions to and from any or all provisions of this chapter as it may
find necessary and proper in the public interest." 29 U.S.C. § 628 (1994). The EEOC
regulations regarding the ADEA are codified at 29 C.F.R. §§ 1625-27 (1998).
129 The ADEA was originally enforced by the Wage and Hour Division of the United
States Department of Labor. In 1978, President Carter, acting pursuant to the Reorganization
Act of 1977, 5 U.S.C. § 901 (1994), transferred ADEA enforcement power to the EEOC. See
Reorg. Plan No. 1 of 1978, supra note 4 (transferring enforcement and regulatory power over
the ADEA from Department of Labor to the EEOC).
130 It would be futile to argue that President Carter exceeded his powers by simply
transferring authority to the EEOC because the United States Constitution has vested in him the
"executive Power." U.S. CoNsT. art. II, § 1, cl. 1. This "Power" would not allow him to grant
rulemaldng authority, but a mere transfer of Congress's delegated power has never been
challenged on constitutional grounds. See supra notes 116-17 and accompanying text
(discussing the principle that Congress, not the President, has the authority to grant rulemaking authority to the EEOC under Title VII).
1999]
JUDICIALDEFERENCETO THEEEOC
1557
ADEA, which comport with APA procedures, should be entitled to compelled
deference. 13 1 As the following section illustrates, EEOC views with respect to
the ADEA have not always received such deference.
1. The Meaning of "Subterfuge " Under the ADEA
Although age-based employment decisions are generally prohibited by the
ADEA, an exemption exists in section 4(2) of the Act for "any bona fide
employee benefit plan... which is not a subterfuge to evade the purposes of' the
Act. 132 The EEOC adopted regulations which provided that plans would qualify
for the exemption (i.e., would not be considered a "subterfuge") only if agerelated reductions in benefits are justified by actuarial data that reveals the
increased cost of providing such benefits to older workers. 133 The Supreme
Court has persistently rejected this interpretation of "subterfuge."
The Supreme Court first adopted its definition of the ADEA term
"subterfuge" in United Air Lines, Inc. v. McMann. 134 McMann addressed the
lawfulness of a mandatory retirement plan adopted prior to enactment of the
ADEA, which turned on the definition of subterfuge in the ADEA safe harbor
provision. The Court adopted the following dictionary definition of subterfuge:
"a scheme, plan, stratagem, or artifice of evasion." 13 5 In the context of section
4(2), the Court reasoned that this definition connotes a specific "intent ... to
131 As a reminder, compelled deference, means that a court should pay deference to
agency interpretations promulgated pursuant to the APA, and grounded in an express
congressional grant; unless they are arbitrary, capricious, or manifestly contrary to the statute.
See supra notes 47-53 and accompanying text (discussing the compelled deference standard).
However, the majority of the Supreme Court does not apply the compelled deference standard
to interpretive rules. See supra note 56 and accompanying text (discussing Scalia's
disagreement with exclusion of interpretive rules from Chevron deference).
132 29 U.S.C. § 623(f)(2) (1994) (emphasis added). This exemption is commonly known
as a "safe harbor" provision. For a discussion of the subterfuge issue in the context of the
ADEA, see James J. Brudney, Congressional Commentary on Judicial Interpretations of
Statutes: Idle Chatteror TellingResponse?, 93 MICH. L. REv. 1, 11-16 (1994).
133 The precise language of the regulations defining "subterfuge" is as follows: "In
general, a plan or plan provision which prescribes lower benefits for older employers on
account of age is not a 'subterfuge' within the meaning of section 4(f)(2), provided that the
lower level of benefits is justified by age-related cost considerations." 29 C.F.R. § 1625.10
(1998). The Department of Labor formulated this rule, and subjected it to notice and comment
in 1969. See 29 C.F.R. § 860.120(a) (1970). After the EEOC received enforcement authority in
1978, it retained the rule in its 1979 regulations, which were also subjected to notice and
comment. See 29 C.F.R. § 1625.10 (1998).
13 4
434 U.S. 192,203 (1977).
135 Iad at 203.
1558
OHIOSTATE LA WJOURNVAL
[Vol. 60:1533
evade a statutory requirement." 136 Because the challenged plan was
implemented before the ADEA was passed, plaintiffs could not logically
1 37
establish that it was specifically intended as a subterfuge to evade the Act.
Accordingly, plaintiffs failed to satisfy the intent requirement in the Court's
definition of subterfuge and were denied any relief.
Congress disagreed with this holding, and amended the ADEA to state that
the bona fide plan exemption did not apply to plans that "require or permit the
involuntary retirement of any individual" because of his or her age.138 However,
Congress did not remove the word subterfuge from the exemption, and thus
paved the way for another controversial Supreme Court decision.
The Court revisited the subterfuge issue eleven years later in Public
Employees Retirement System of Ohio v. Betts.139 The Court was faced with the
question of adopting the McMann definition of subterfuge and its implied intent
requirement or deferring to the EEOC cost justification rule. The question arose
from the following facts. Ohio adopted a retirement plan for its public sector
employees that did not pay disability benefits for employees retiring after age
fifty-nine. 14 0 Under this plan, June Betts, who retired with a disability at age
sixty-one, could receive standard retirement benefits but not disability benefits
14 1
(which were much higher).
Betts filed an age discrimination charge against Public Employment
Retirement System (PERS) with the EEOC, and filed suit in the United States
District Court for the Southern District of Ohio, claiming that PERS' refusal to
pay disability retirement benefits violated the ADEA. 142 The Supreme Court
granted certiorari after two lower court decisions favoring Betts. 143 It held that
136
1d
137
See id. ("[A] plan established in 1941, if bona fide,... cannot be a subterfuge to
evade an Act passed 26 years later.").
138 Age Discrimination in Employment Act Amendments of 1978, Pub. L. 95-256,
§§ 2(a), 623 (f)(2), 92 Stat. 189, 189 (1978) (amending the Age Discrimination in Employment
Act of 1967,29 U.S.C. § 623(f(2)).
139 492 U.S. 158, 169 (1989).
140
See Betts, 492 at 158.
141 The Ohio plan was amended in 1976 to establish a floor amount for disability benefits
at 30% of the retiree's final average salary. See id. at 163. Consequently, the disability
retirement plan would have paid Betts $355 per month, while the standard retirement payments
(referred to as "age-and-service" payments) yielded only $158.80 per month. See id
142
See id. at 163-64.
143 Both lower courts in Betts deferred to the EEOC view, and concluded that PERS was
not entitled to the bona fide plan exemption. See Betts v. Hamilton County Bd. of Mental
Retardation, 631 F. Supp. 1198, 1202-05 (S.D. Ohio 1986), affld, Betts v. Hamilton County
Bd. of Mental Retardation and Dev. Disabilities, 848 F.2d 695 (6th Cir. 1988).
1999]
JUDICIAL DEFERENCETO THE EEOC
1559
the term subterfuge must be given its "ordinary meaning" (or dictionary
definition) as assigned in McMann.14 4 Because this definition "includes a
subjective [intent] element that the regulation's objective cost-justification
requirement fails to acknowledge," the Court rejected the EEOC rule as
'145
"inconsistent[ ] with the plain language of the statute.
The fate of the EEOC guidelines in Betts, like many other Supreme Court
deference cases, was determined by the first step of Chevron-the plain meaning
test. 14 6 However, this is a questionable interpretation of Chevron. First,Chevron
provides that, if Congress-(notthe Supreme Court or Webster's Dictionary) has
spoken clearly to the issue, then reference to agency guidance is not
appropriate. 147 Congress did not speak clearly to the meaning of subterfuge; it
only listed the word. In essence, the Court in Betts found that the EEOC's
position was inconsistent with a "subjective element," which was inferred from a
dictionary definition of subterfuge. 148 This subordination of EEOC guidance in
the process of statutory interpretation is hardly a proper application of Chevron's
149
plain meaning rule.
The second flaw in the Court's reasoning arises from the Court's
examination of legislative history, even after declaring that the plain statutory
meaning precluded agency deference. 150 After rejecting EEOC guidance, the
Court turned to the legislative history to solve the "difficult task of determining
144 "As Congress did not amend the relevant statutory language, we see no reason to
depart from our holding in McMann that the term 'subterfuge' is to be given its ordinary
meaning ... "Betts, 492 U.S. at 168.
14 5
Id. at 171.
14 6
See Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837,
842 (1984). "[N]o deference is due to agency interpretations at odds with the plain language of
the statute itself .... Even contemporaneous and longstanding agency interpretations must fall
to the extent they conflict with statutory language." Betts, 492 U.S. at 171.
147 See Chevron, 467 U.S. at 842-43.
148 Betts, 492 U.S. at 171.
149 Professor James J. Brudney, in his critical analysis of Betts, focuses on the Court's
disregard of legislative history. He noted that a conference report accompanying the 1978
amendment to the ADEA that explicitly disapproves of the McMann definition of subterfuge
'vas deemed inconsequential because Congress had failed to amend the subterfuge tet." See
Brudney, supranote 132, at 15. Thus, the Court not only subordinated EEOC guidance, but
also a clear statement of intent from ADEA legislative history.
150 Application of the plain meaning doctrine precludes not only inquiry into agency
interpretation but also legislative history. See Danial A. Farber & Phillip P. Frickey, Legislative
Intent and Public Choice, 74 VA. L. REV. 423, 438-39 (1988) (quoting then Judge Scalia's
criticism of legislative history and its "system of committee-staff prescription" in Hirschey v.
Federal Energy Regulatory Commission, 777 F.2d 1, 7-8 (D.C. Cir. 1985) (Scalia, J.,
concurring)).
1560
OHIO STATE LA WJOURNAL
[Vol. 60:1533
the precise meaning of [subterfuge], 1 5 1 and ultimately found conflict between the
EEOC view and Congress's intent as expressed in legislative history. However,
if the plain meaning of subterfuge was clear enough to justify rejection of the
EEOC view, then the court should not need to refer to legislative history.
Justice Marshall, in dissent, recognized that the statutory language was not
clear, and turned to sources such as legislative history and agency guidance to
resolve the ambiguity.15 2 After concluding that the EEOC guidelines were
consistent with ADEA legislative history, Justice Marshall argued that the EEOC
position was entitled to deference. 153 He then boldly accused the Court of
inconsistent application of its deference standard when "important civil rights
laws are at issue .... -154
2. CongressionalOverride of Betts
On October 16, 1990, President Bush signed into law the Older Workers
Benefit Protection Act of 1990 (OWBPA).15 5 Title I of the OWBPA amended
the ADEA for the purpose of overruling the Betts decision. 156 In light of the
OWBPA, section 4(2) of the ADEA now permits use of the bona fide benefit
plan exemption only when "the actual amount of payment made or cost incurred
on behalf of an older worker is no less than that made or incurred on behalf of a
younger worker .... -157 This amendment comports with EEOC guidelines
which were disregarded in the Betts decision. 15 8 The OWBPA also made clear
that the amended ADEA subsection 4(f)(2) is an affirmative defense which shall
apply regardless of the date when the disputed benefit plan was adopted
(overruling McMann), and as to which the employer shall have the burden of
See Betts, 492 U.S. at 175.
Justice Marshall noted that "[t]his is a case in which only so much blood can be
squeezed from the textual stone, and in which one therefore must turn to other sources of
151
152
statutory meaning." Id. at 187 (Marshall, J., dissenting).
153 See id. at 192-93 (Marshall, J., dissenting).
154
Id. at 193 (Marshall, J., dissenting). The precise language of his accusation was: "The
majority's derogation of [the EEOC guidelines] leaves one to wonder why, when important
civil rights laws are at issue, the Court fails to adhere with consistency to its so often espoused
policy of deferring to expert agencyjudgment on ambiguous statutory questions." Id. Earlier in
his dissenting opinion, Justice Marshall also commented that the majority's reasoning is "so
manipulative as virtually to invite the charge of result-orientation." Id. at 185.
155 Older Workers Benefit Protection Act of 1990, Pub. L. No. 101-433, 104 Stat. 97884 (codified as amended at 29 U.S.C. §§ 621,623, 626, 630 (1994 & Supp. M 1997).
156 See John R. Runyan, HedgingBetts: The Older Workers Benefit ProtectionAct, 72
MICH. B., Feb. 1993, at 168, 171 (discussing Congressional override of Betts).
157 29 U.S.C. § 623(f(2)(B) (1994).
158 See 29 C.F.R. § 1625.10 (1998) (articulating the cost-justification rule).
1999]
JUDICIAL DEFERENCETO THE EEOC
1561
9
proof (overruling Betts). 15
One lesson to be learned from the OWBPA is that the entire effort would
have been unnecessary had the Court given proper deference to EEOC guidance.
It is simply wasteful to require Congress to spend time redressing an issue that it
160
originally intended the agency to resolve.
It appears that courts interpreting the ADEA have observed the
congressional mandate embodied in the OWBPA. 161 However, when the same
issue arose under the ADA, 162 many courts followed the Betts decision. At
present a split exists in the lower courts as to the validity of EEOC guidelines
interpreting the ADA term subterfuge. The following sections explore this issue
(perhaps the most contentious ADA deference issue in the courts today), and
other problems associated with EEOC interpretations of the ADA.
C. Deference to EEOCInterpretationsofthe ADA
Because of its youth, the ADA and its EEOC guidelines present many
difficult interpretive problems for the courts. The following sections focus on
two such problems. First the "mitigating measures" rule, which the EEOC has
adopted to clarify the meaning of the ADA phrase "substantial limitation," had
caused a major split in lower federal courts. The split was nearly resolved by the
circuit courts in favor of deference to the rule. However, the Supreme Court in
Sutton v. UnitedAirlines,Inc.,163 recently rejected the EEOC view with regard to
164
mitigating measures.
The primary reason for a discussion of the mitigating measures rule is to
provide context for a second, more contentious issue: the meaning of subterfuge
under the ADA. Very few circuit courts have addressed this issue, but a clear
split is emerging.
15 9 See
Runyan, supra note 156 (discussing affirmative defense and burden of proof
amendments to Subsection 4(f(2) of the ADEA).
160 Professor James J. Brudney points out that the process of overruling Betts was a
lengthy and controversial one. See Brudney, supranote 132, at 17-19.
16 1
Numerous courts have recognized that Betts has been superseded by the OWBPA.
See, e.g., EEOC v. Mass., 77 F.3d 572,573 (1st Cir. 1996); American Ass'n of Retired Persons
v. Farmers Group, Inc., 943 F.2d 996 (9th Cir. 1991); EEOC v. Westinghouse Elec. Corp., 925
F.2d 619 (3d Cir. 1991).
162 For better or worse, Congress used the same term in the ADA in a similar "safe
harbor" provision. See infra Part IV.C.2 (discussing the subterfuge issue in the context of the
ADA).
163 119 S.Ct. 2139 (1999).
164
See id at 2151.
1562
OHIOSTATE LAWJOURNAL
[Vol. 60:1533
1. JudicialDeference to the MitigatingMeasuresRule
The mitigating measures rule is located in the EEOC "Interpretive
Guidance" as an appendix to the ADA regulations. 165 The EEOC has clear
congressional authority under the ADA (unlike Title VII) to issue regulations
with the force of law 166 and such regulations undoubtedly trigger Chevron's
compelled deference standard. As to interpretive guidance, the deference
standard is less clear. Both the form and substance of the guidance will be
relevant to its ultimate categorization.
In form, this "Interpretive Guidance" does not appear to qualify for
compelled deference for the simple reason that it is not a "regulation."' 167
However, numerous factors cast doubt on this proposition. First the label of an
agency rule, although relevant, is not always dispositive as to its substance.1 68
Second, the Guidance was set forth as an appendix to the regulations-not as a
separate body of rules. 169 The existence of an appendix merely indicates that an
agency is interpreting its own regulations, not that the agency is imposing
entirely new rules in addition to the regulations. In fact, the D.C. Circuit has
recognized that an agency interpretation of its own regulations is entitled to more
weight than its interpretation of an ambiguous statutory term. 170 Therefore, the
fact that the guidance is in appendix form may allow an inference, albeit weak,
that the EEOC was attempting to exercise its delegated rulemaking authority. 171
165 See 29
166
C.F.R. 1630, app. § 1630.2(h) (1998).
See 42 U.S.C. 12116 (1994) (authorizing the EEOC to issue guidelines implementing
Title I of the ADA).
167 Recall that Chevron only applies to "legislative regulations." See supraPart ll.B.
16 8
See General Motors Corp. v. Ruckelshaus, 742 F.2d 1561, 1565-66 (D.C. Cir. 1984)
('The agency's own label, while relevant, is not dispositive .... [I]f by its action the agency
intends to create new law, rights or duties, the rule is properly considered to be a legislative
rule.") (citations omitted).
169 The regulations and interpretive guidance share corresponding citations. For example,
the regulation "28 C.F.R. § 1630" will correspond to '28 C.F.R. § 1630, app." in the
interpretive guidance. Each section of Interpretive Guidance simply expounds on its
corresponding regulation.
170
See Birt v. Surface Trans. Bd., 90 F.3d 580, 589 (D.C. Cir. 1996) (stating that
"precedent firmly establishes that courts' owe even greater deference to agency interpretations
of agency rules than ... to agency interpretations of ambiguous statutory terms.") (quoting
Capital Network Sys., Inc. v. FCC, 28 F.3d 201,206 (D.C. Cir. 1994)).
171 The fact that the guidance is in appendix form lead to a contrary inference in Matzak
v. FranfordCandy & Chocolate Co., 136 F.3d 933 (3d Cir. 1997). The court noted that 'The
EEOC's guidelines constitute an appendix to the regulations and therefore do not command the
same degree of deference as the regulations themselves.' Id. at 937. The court in Matzak
ultimately decided to give the EEOC interpretive rule 'controlling weight unless it is plainly
1999]
JUDICIAL DEFERENCETO THE EEOC
1563
A third reason for placing the interpretive guidance into the legislative
172
category is that it was subjected to notice and comment pursuant to the APA.
This is probably the most compelling argument for treating EEOC interpretive
guidance as a body of legislative rules, because the agency was not obligatedto
follow that procedure. 17 3 The fact that the EEOC followed the notice and
comment procedure for its Interpretive Guidance suggests that it was attempting
to exercise its delegated rulemaking authority.
Even if the guidance is deemed "interpretive" in form, its substance may
suggest otherwise. 174 The substantive value of a rule may turn on the impact of
the rule upon the rights and duties of individuals. 175 The impact of the mitigating
measures rule would be abstract if discussed here. 176
erroneous or inconsistent with the regulations."'Id (citing Thomas Jefferson Univ. v. Shahala,
512 U.S. 504,512 (1994)).
172
Section 553 of the APA sets forth the notice and comment requirements. Under this
procedure, informal rules must be published in the Federal Register (the notice element) so that
interested parties or entities can offer criticism or suggestions (the comment element). See 5
U.S.C. §§ 553(b), 553(c) (1998). On August 1, 1990, the EEOC issued advance notice of its
proposed rule regarding the definition of ADA terms such as "disability," "reasonable
accommodation," and "undue hardship." See Title I of Americans with Disabilities Act;
Implementation, 55 Fed. Reg. 31192 (1990) (to be codified at 29 C.F.R. pt. 1630)(advanced
notice of proposed rulemaldng Aug. 1, 1990). Six months later, it issued its final notice of
proposed rulemaking which included both the regulations and interpretive guidance. See Equal
Employment Opportunity for Individuals with Disabilities; Notice of Proposed Rulemaking,
56 Fed. Reg. 8586-8603 (1991) (to be codified at 29 C.F.R. pt 1630) (proposed Feb. 28,
1991). On July 26, 1991, after an extensive comment and amendment process, the EEOC
published the final version of its rules. See Equal Employment Opportunity for Individuals
with Disabilities, 56 Fed. Reg. 35726-56 (1991) (to be codified at 29 C.F.R. pt. 1630) (final
rule).
173 Interpretive rules are categorically exempt from the APA notice and comment
requirement. See 5 U.S.C. § 553(b)(A) (1994).
174
Recall the debate discussed above regarding the characterization of 'legislative" or
"interpretive" rules. Some cases and scholars focus on the origin of the rule (e.g., its
nomenclature and form), while others focus on the impact of the rule. This debate is entirely
relevant to ADA interpretive guidance, which in form appears to be interpretive, but in
substance, resembles a legislative rule. See supra Part II.C. (discussing the distinction between
legislative and interpretative rules).
175
See Chrysler Corp. v. Brown, 441 U.S. 281,302 (1979) ("We described a substantive
rule---or a 'legislative-type rule,'-as one 'affecting individual rights and obligations' ... This
characteristic is an important touchstone for distinguishing those rules that may be 'binding' or
have the 'force of law.") (quoting Morton v. Ruiz, 415 U.S. 199,236 (1974)).
176 However academic the distinction between legislative and interpretive rules may
seem, it may be critical in a deference context. If the guidance is deemed legislative, then it
almost automatically triggers the compelled deference standard. If not, the earned deference
standard would apply. See supra Parts IIHA and II.B (discussing, respectively, earned and
1564
OHIO STATE LAWJOURNAL
[Vol. 60:1533
The mitigating measures rule is used both to determine whether an
individual's physiological disorder constitutes a physical impairment, and
whether that impairment substantially limits one or more major life activities of
the individual. As to whether a person is impaired under the ADA, the
interpretive guidance states that "[t]he existence of an impairment is to be
determined without regard to mitigating measures such as medicines, or assistive
or prosthetic devices." 177 A similar rule applies as to whether the impairment
satisfies the substantial limitation requirement: "The determination of whether an
individual is substantially limited in a major life activity must be made on a case
by case basis, without regard to mitigating measures ....
"178 Therefore, under
the EEOC rule, a physiological disorder will be considered an impairment, and
the impairment will be substantially limiting, even if mitigating measures can
1 79
completely control the disorder.
One example of nearly complete mitigation is an insulin-dependent diabetic
who regularly takes insulin. Without the measures, the person would likely lapse
into a coma and die. With the measures, he or she leads a normal, and seemingly
"unlimited" lifestyle. 180 Under the EEOC rule, however, the person is still
considered "disabled," because it mandates that insulin cannot be considered in
the disability equation. This result is certainly controversial and much ink has
been spilled on its policy implications. 181 This Note does not explore those
implications in depth, 182 but does focus on the level of deference paid to the rule
compelled deference standards).
177
29 C.F.R. § 1630.2(h), app. (1999).
app. (1999).
178 29 C.F.R. § 1630.2(j),
17 9 Numerous
sources of ADA legislative history provide clear support for this view. See
H.R. Rep. No. 101-485(1), at 53, reprintedin 1990 U.S.CAA.N. 303, 334 ("Whether a
person has a disability should be assessed without regard to the availability of mitigating
measures ....") (House Education and Labor Committee). The other two relevant committee
reports contain nearly identical language. See H.R. Rep. No. 101-485(m1), at 28-29 (1990),
reprintedin 1990 U.S.C.A.A.N. 445, 451 (Judiciary Committee); S.Rep. No. 101-116, at 23
(1989) (Labor and Human Resources Comrnmittee). Nothing in the legislative history suggests
any disagreement with this position.
180 'Memitigated diabetic issue has prompted numerous ADA cases. See, e.g., Arnold v.
United Parcel Serv., Inc., 136 F.3d 854, 863-64 (1st Cir. 1998) (granting standing to an
insulin-dependent diabetic, and reversing the lower court decision to disregard mitigating
measures rule); Gilday v. Mecosta County, 124 F.3d 760 (6th Cir. 1997) (reversing and
remanding over the question of whether non insulin-dependent diabetic was disabled).
181 See, e.g., Recent Cases, Statutory Interpretation-Americans with Disabilities
Act-Tenth Circuit holds that Courts Should Consider Mitigating Measures in Evaluating
Disability.-Sutton v. UnitedAir Lines, Inc., 130 F.3d 893 (10th Cir. 1997), 111 HARV.L.
REv.2456,2460-61 (1998).
182
One logical policy argument in favor of the rule is that, without it, employers could
1999]
JUDICIAL DEFERENCETO THE EEOC
1565
both in the past and present.
One procedural problem with the mitigating measures rule, which
surprisingly has not yet been mentioned by any appellate courts addressing its
validity (even the ones that rejected it),183 is the fact that the EEOC's original
notice publication only included the rule as it applied to impairments, not to
substantial limitation. 184 The rule was not included in the substantial limitation
section of the guidance until the final draft. 185 Thus, from a strict procedural
perspective, the mitigating measures rule with regard to "substantially limits" has
failed the APA notice requirement even though it was amended in direct
186
response to the comment process.
Although the notice draft did not explicitly mention mitigating measures, it
did contain the following example: "A diabetic who without insulin would lapse
into a coma would be substantially limited because the individual can only
perform major life activities with the aid of medication." 187 As one scholar
points out this language implied that the consideration of whether someone is
188
substantially limited should be made without regard to mitigating measures.
institute treatment in order to remove individuals from the ADA protection. Once the employee
is cured, the employer could fire him or her with impunity. One logical argument against the
rule is the fact that it could open the floodgates for ADA plaintiffs who experience no
debilitating symptoms of their conditions.
183 One appellate court appeared to understate the procedural soundness of the
interpretive guidance. See Arnold, 136 F.3d at 864 ("We recognize that the EEOC interpretive
guidelines are not controlling in the way that regulations promulgated pursuant to the
Administrative Procedure Act are controlling.') (citing 5 U.S.C. § 552 (1994 & Supp. IV
1998)). Even though the court ultimately accepted the EEOC position, it did not mention the
fact that the interpretive guidance was subjected to notice and comment pursuant to Section
553 of the APAee 5 U.S.C. § 553(b) (1994).
184
See Equal Employment Opportunity for Individuals with Disabilities; Notice of
Proposed Rulemaking, 56 Fed. Reg. 8586-8603 (1991) (to be codified at 29 C.F.R. pt. 1630)
(proposed Feb. 28, 1991) (stating mitigating measures rule with regard to impairments).
185 "The Commission has revised the interpretive guidance accompanying § 1630.2G) to
make clear that the determination of whether an impairment substantially limits one or more
major life activities is to be made without regard to... mitigating measures." See Equal
Employment Opportunity for Individuals with Disabilities, 56 Fed. Reg. 35,726, 35,727 (1991)
(to be codified at 29 C.F.R. pt. 1630) (final rule).
186
The EEOC states that the amendments to the substantial limitation section were in
response to comments by disability rights groups that requested clarification of the substantial
limitation requirement. See idat 35,728.
187 Equal Employment Opportunity for Individuals with Disabilities; Notice of Proposed
Rulemaking, 56 Fed. Reg. at 8593.
188
See Colker, supranote 11, at 154-55.
1566
OHIO STATE LA WJOURNAL
[Vol. 60:1533
The issue of deference to the mitigating measures rule created a great rift in
the district courts, but the circuit courts appeared much more deferential. 189 The
issue has been addressed in a total of sixteen appellate cases. 190 Of those cases,
only three resulted in a rejection of the mitigating measures rule. 191 Of the
189 Most
of the district court cases represented in my sample resulted in rejection of the
mitigating measures rule. Of the 13 circuit court cases in my sample resulting in acceptance of
the rule, 9 of those cases were reversals of a district court decision to reject the rule; and all 3 of
the circuit court cases resulting in rejection of the rule were afirmances of a similar district
court decision. Therefore, in the vast majority (approximately 75%) of district court decisions
that were reviewed in my sample of appellate cases, the result was a rejection of the mitigating
measures rule. Although numerous other district court cases not represented in my appellate
sample have considered the mitigating measures issue, one can safely infer from this data that
district courts were generally more hostile to the rule than circuit courts (at least until they were
reversed). I did review the entire district court record briefly to confirm this conclusion. From
this admittedly cursory search, it appeared that most district courts did reject the rule, although
it seemed quite wasteful to conduct a detailed empirical analysis of those cases, especially since
most of the those cases are no longer good law.
190 To gather the initial cite list, I searched for "mitigating measures" in the WESTLAW
ALLCIR database in early January of 1999. I excluded "unpublished" opinions from my
sample. The reason for this exclusion is not only because of the weak precedential value of
those opinions, but also because certain circuits make their opinions available and others do
not. Thus, if I included all unpublished decisions, then the views of one circuit that
disseminates its unpublished decisions to electronic sources (e.g., the Sixth Circuit), would be
more represented in my sample than a circuit that does not (e.g., the Third Circuit). In light of
these problems, I excluded unpublished decisions altogether. It is important to note that
unpublished opinions are valuable for statistical analysis, especially when determining legal
trends from a relatively small sample of published opinions. Some scholars are quite interested
in obtaining unpublished opinions for their research, or at least, in determining how many
opinions they cannot obtain in order to speculate on rate of error. In a study I conducted on
behalf of Professor Ruth Colker at The Ohio State University College of Law in May of 1998,
I examined the practices of all circuit courts regarding the availability of their unpublished
opinions. I will not subject the reader to the results of that study, but Professor Colker does
summarize them in her most recent article. See Colker, supra note 11 at nn. 26-38 & app. A.
For a complete list of the cases resulting in both acceptance and rejection of the mitigating
measures rule, see apps. C & D.
191 The following cases resulted in rejection: Sutton v. United Airlines, Inc., 130 F.3d
893, 899 n.3 (10th Cir. 1997) ("[A]lthough we give great deference to the EEOC's
interpretation of the ADA found in the regulations promulgated under the express authority of
Congress and the ADA itself, we do not do the same for interpretive guidance promulgated
under the [APA].... Hence, while the EEOC's Interpretive Guidance may be entitled to some
consideration in our analysis, it does not carry the force of law and is not entitled to any special
deference [under Chevron]'); Gilday v. Mecosta County, 124 F.3d 760, 766 (6th Cir. 1997)
(Kennedy, J., concurring) ("[T]he EEOC's rule on mitigating circumstances conflicts with the
text of the ADA and is, therefore, not a 'permissible construction of the statute."') (citing Reno
v. Koray, 515 U.S. 50, 61 (1995)); Ellison v. Software Spectrum, Inc., 85 F.3d 187, 191 n3
(5th Cir. 1996) ("[H]ad Congress intended that substantial limitation be detemined without
1999]
JUDICLIL DEFERENCETO THE EEOC
1567
thirteen cases in which the rule was not rejected, the amount of deference paid to
1 93
1 92
the rule varied widely from "controlling weight," to "considerable weight,"'
to "some weight."'194 In many of these thirteen cases, the courts did not explicitly
address the issue of deference; rather, they merely cited the EEOC rule and
applied facts of the case directly to that rule. 195 These cases may arguably exhibit
the highest possible deferential standard because they seem to accept EEOC
19 6
guidance without question or controversy.
Despite the overwhelming number of appellate cases accepting the EEOC
view, the Supreme Court went the other way. In Sutton v. United Airlines,
Inc., 197 the Court determined that the EEOC view with regard to mitigating
measures "is an impermissible interpretation of the ADA."'1 98 Sutton involved
two severely myopic job applicants who were denied employment as United
Airlines pilots because they did not meet United's minimum vision
requirements. 199 Their vision was entirely correctable, but under the EEOC rule,
a court must determine disability status without regard to any corrective or
mitigating measures. It seems intuitive that the mitigating measures rule should
not apply to simple myopia. However, the Court did more than deactivate the
regard to mitigating measures, it would have provided for coverage under [42 U.S.C.
§ 12102(2)(A) (1998)] for impairments that have thepotentialto substantially limit a major life
activity.").
192
Matzak v. Frankford Candy & Chocolate Co., 136 F.3d 933, 937 (3d Cir. 1997)
("[W]e must give the EEOC's interpretation of its own regulations 'controlling weight unless it
is plainly erroneous or inconsistent with the regulation[s]."') (citing Thomas Jefferson Univ. v.
Shahala, 512 U.S. 504,512 (1994)).
193 Harris v. H & W Contracting Co., 102 F3d 516,521 (11th Cir. 1996) ('The Supreme
Court has long recognized that an agency's interpretation of a statute it is entrusted to
administer should be given 'considerable weight' and should not be disturbed unless it appears
from the statute or legislative history that Congress intended otherwise.) (citing Chevron
U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837,844-45 (1984)).
194 Arnold v. United Parcel Serv., Inc., 136 F.3d 854, 864 (1st Cir. 1998) ("[EEOC
interpretive guidelines] deserve at least as much consideration as a mere 'internal agency
guideline,' which the Supreme Court has held is entitled to 'some deference' as long as it is a
permissible construction of the statute.) (citing Reno v. Koray, 515 U.S. 50, 61 (1995)).
195 See, eg., Baert v. Euclid Beverage, Ltd., 149 F.3d 626, 629-31 (7th Cir. 1998);
Holihan v. Lucky Stores, Inc., 87 F3d 362, 366 (9th Cir. 1996); Burch v. Coca-Cola Co., 119
F.3d 305,317 (5th Cir. 1997); Foreman v. Babcock & Wilcox Co., 117 F.3d 800,805-06 (5th
Cir. 1997).
196 Logically it would be more deferential for a court to simply cite and apply an EEOC
rule, than to engage in a lengthy and often confusing deference analysis, which could be
misconstrued or disagreed with by future litigants or courts.
197 119 S. Ct. 2139 (1999).
198 Id at 2146.
199 See id at 2143.
1568
OHIO STATE LA WJOURNAL
[Vol. 60:1533
EEOC rule within the context of myopia. It entirely invalidated the mitigating
measures rule, regardless of the impairment to which it applies. This blunt
measure adopted by the Supreme Court could affect the level of deference paid
to the EEOC "subterfuge" rule.
2. "Subterfuge" Underthe ADA
The ADA subterfuge issue is complex and requires careful explanation.2 00
The applicability of Title I to employer-provided health insurance is affected by
two incompatible goals: the elimination of discrimination based on disability,
and the preservation of affordable employee health insurance plans. 2 01 To
accomplish the first goal, Congress enacted Title I of the ADA, which prohibits
discrimination in all aspects of employment.20 2 Subsequent EEOC regulations
extended that protection to the broad category of "fringe benefits,"2 0 3 however,
the EEOC limits the application of its interim guidance to only employerprovided health insurance plans.20 4 Thus, through a combination of statutory and
regulatory action, the ADA prohibits disability-based distinctions in employerprovided health plans.
Congress attempted to accomplish the second goal of preserving affordable
health insurance through the creation of a statutory exemption (or "safe harbor"
provision) for certain distinctions in employer-provided health plans.205 Section
200 For
a brief overview of the issue, see Brudney, supranote 132, at 199 n.395.
201 See CCR REPORT ON ADA TrLE I, supranote 1, at 134.
202
See 42 U.S.C. 12112(a) (1994) ("No covered entity shall discriminate against a
qualified individual with a disability because of the disability of such individual in regard to job
application procedures, the hiring, advancement, or discharge of employees, employee
compensation, job training, and other terms, conditions, and privileges of employment.").
203 See 29 C.F.R § 1630A(f) (1998) ("It is unlawful for a covered entity to discriminate
on the basis of disability against a qualified individual with a disability in regard to: ...(f)
Fringe benefits available by virtue of employment, whether or not administered by the covered
entity.').
204
See Interim Enforcement Guidance on the Application of the Americans With
Disabilities Act of 1990 to Disability-Based Distinctions in Employer-Provided Health
Insurance, EEOC Coml. Man. (BNA) No. N-915.002, at 2 (June 8, 1993) [hereinafter EEOC
Interim Guidance on Health Insurance]. The fact that the EEOC voluntarily limited the scope
of its guidance could increase the inherent weight of the guidance in the eyes of a court,
because this suggests that EEOC fact-finding and deliberation was narrowly focused toward a
discrete issue.
205 This exemption was a natural response to efforts by insurers to cut costs by
differentiating between, and offering disparate levels for, particular health conditions. See H.
Miriam Farber, Note, Subterfuge: Do Coverage Limitations and Exclusions in EmployerProvidedHealth Care Plans Violate the Americans with DisabilitiesAct?, 69 N.Y.U. L. REV.
850, 851-52 (1994).
1999]
JUDICIALDEFERENCETO THEEEOC
1569
501(c) of the ADA exempts certain "bona fide" benefit and insurance plans from
the general restrictions of the Act.20 6 However, employers or entities cannot
receive safe harbor when their plans are used as a "subterfuge to evade the
20 7
purposes" of the ADA.
Again, the obvious problem arising from the section 501(c) provision is that
the meaning of subterfuge is not self-evident. As the Sixth Circuit once noted,
section 501(c) is "totally ambiguous on its face." 20 8 This ambiguity necessarily
leads a reviewing court to outside sources to ascertain the proper meaning of
subterfuge. Therefore, the question in ADA subterfuge cases becomes: what is
the best outside authority on the meaning of or congressional intent behind the
word subterfuge? Is it Congress, speaking through legislative history; the
206 Section 501(c), under the title "Insm-ance" provides:
Titles I through IV of this Act shall not be construed to prohibit or restrict(1) an insurer, hospital or medical service company, health maintenance
organization, or any agent, or entity that administers benefit plans, or similar organizations
from underwriting risks, classifying risks, or administering such risks that are based on or
not inconsistent with State law; or
(2) a person or organization covered by this chapter from establishing, sponsoring,
observing or administering the terms of a bona fide benefit plan that are based on
underwriting risks, classifying risks, or administering such risks that are based on or not
inconsistent with State law, or
(3) a person or organization covered by this chapter from establishing, sponsoring,
observing or administering the terns of a bona fide benefit plan that is not subject to State
laws that regulate insurance.
42 U.S.C. § 12201(c) (1994).
2 07
Id. ("Pargraphs (1), (2), and (3) shall not be used as a subterfuge to evade the
purposes of subchapter I and IIm of this chapter.'). The health plan provider has the burden of
proving that the plan is not being used as a subterfuge. See EEOC Interim Guidance on Health
Insurance, supranote 204, at 5.
208 Parker v. Metropolitan Life Ins. Co., 99 F.3d 181, 190 (6th Cir. 1997). In Parker,
Judge Gilbert Merritt expressed considerable frustration over the ambiguities inherent in
Section 501(c):
Unlike the language of Title m11,
which is quite clear, the meaning of the "safe
harbor" provision is not self-evident In fact, the statute appears to be purposefully vague
in order to satisfy contending interest groups. Unable to decide on exactly what it intended
to legislate, Congress inserted language which looks in both directions. One provision
attempts to appease the insurance industry; the other provisions attempt to help the large
group of disabled people. In doing so, Congress has again left this Court in the position to
give meaning to conflicting statutory language designed as a political compromise. We
find that, in this instance, the statute is totally ambiguous on its face.
Id at 190.
1570
OHIO STATELAWJOURNVAL
[Vol. 60:1533
Supreme Court, speaking in an overruled ADEA case; or the EEOC, speaking
through its interim enforcement guidance?
a. CongressionalIntent as to the MeaningofSubterfuge
The ADA legislative history could hardly be more clear as to Congress's
intent regarding subterfuge.2 09 According to unequivocal statements from
2 11
Edwards 2 12 and Senator Kennedy, 213
Representatives Waxman,210 Owens,
the term subterfuge in the ADA does not mean what Betts says it does. The
credibility of these statements is almost unshakable because the ADA was a
bipartisan bill, the speakers were all supporters of the bill, and three of the
statements-those of Mazoli, Kennedy, and Edwards-came from a July 13
conference report, just thirteen days before the bill was passed by an
214
overwhelming vote.
209 This Note only examines explicit remarks by legislators regarding the meaning of
subterfuge. The broader legislative origins of Section 501(c) are intriguing, but slightly beyond
the scope of this Note. For a thorough analysis of those origins, see Farber, supranote 205, at
876-80.
210 "i have been informed by those Members who are closely involved in the legislation
to overturn the decision of the Supreme Court in Public Employees Retirement System of Ohio
v. Betts, 109 S. Ct. 256 (1989), that the term 'subterfuge' in the ADA should not be read as the
Supreme Court read that term in Betts. Thus, there is no requirement of an intent standard
under the ADA.... " 136 CONG. REC. H4626 (daily ed. July 12, 1990) (statement of Rep.
Waman).
211 "Questions have been raised to me recently as to whether the term 'subterfuge' should
be interpreted consistent with the Supreme Court's decision in [Betts]. The answer is 'No.' We
use the term 'subterfuge' as a means of evading the purposes of the ADA. It does not mean that
there must be some malicious or purposeful intent to evade the ADA on the part of the
insurance company or organization." 136 CONG. REC. H4623 (daily ed. July 12, 1990)
(statement of Rep. Owens).
2 12
See 136 CONG. REC. H4624 (daily ed. July 12, 1980) (statement of Rep. Edwards).
Representative Edward's statement was nearly identical to that of Senator Kennedy. See infra
note 213.
213 Senator Kennedy stated:
The provision regarding subterfuge in Section 501(c) [of the ADA] should not be
undermined by a restrictive reading of the term "subterfuge," as the Supreme Court did in
Betts. Indeed, our committee recently reported out a bill to overturn the Betts decision. It
is not our intent that the restrictive reading ofBetts, with which we do not agree, should be
carried over to the ADA [from the ADEA].
136 CONG. REC. S9697 (daily ed. July 13, 1990) (statement of Sen. Kennedy).
214 These factors tend to soften some of the criticisms of legislative history, which include
concerns about the under-representative nature of legislative history as opposed to the statute
1999]
JUDICIALDEFERENCETO THE EEOC
1571
One probing question arises from the above discussion of legislative history.
If Congress was so intent on overruling Betts, why did it not say so in the ADA?
This, along with the timing of the ADA, 215 has fueled the theory that the statute
does embody the Betts interpretation, regardless of the statements of a group of
legislators. This theory is grounded by two factors. First, because Congress was
aware of the Betts interpretation but did not overturn it in the statute, it is
presumed to have acquiesced to that interpretation. 2 16 Second, the OWBPA bills,
which removed the subterfuge language from the ADEA, were introduced in
Congress on August 3 and 4, 1989 2 17-only two days after the subterfuge
language was added to the Senate Labor Committee's ADA draft.2 18 The House
ultimately accepted the Senate version, and the bill was enacted in July of 1990
with the subterfuge language intact The OWBPA was passed on October 16,
1990.219 As one commentator notes, "[t]his timing demonstrates that Congress
was aware of the prior judicial construction of 'subterfuge' in an analogous
statute, knew how to foreclose that construction, and decided nonetheless to add
220
that language to the ADA."
One response to the "acquiescence" argument is that the notion of legislative
acquiescence is only a canon of statutory interpretation that has limited
persuasive value in the courts. 22 1 As to the timing argument, the fact that
itself, and include concerns about legislative sabotage that affect later interpretations of a
statute. For a comprehensive discussion of the theoretical, constitutional, and practical
arguments for disregarding legislative history, see Brudney, supranote 132, at 41-66.
2 15 The account that follows regarding the timing of the ADA is drawn from Farber,
supra note 205, at 895-97. For a similar account, see David A. Copus & Glen D. Nager,
Benefit Plan LimitationsAfter the Americans with DisabilitiesAct, 19 EMPLOYEE REL. L.J.77,
81 (1993). But cf Ronald S. Cooper, EEOC Issues Guidance on Applying the ADA to Health
Insurance Plans, 2 EMPL.
TESTING,
L. & POLY. REP. 125, 128-29 (1993) (taking a more
cautious approach toward the timing issue).
2 16 Professors Eskridge and Frickey describe the "acquiescence rule" as follows: "If
Congress is aware of an authoritative agency or judicial interpretation of a statute and doesn't
amend the statute, Congress is presumed to 'acquiesce' in the interpretation's correctness:' See
ESKRIDGE & FRICKEY, supranote 93, at 814.
2 17
See S. 1511, 101st Cong. (1989); H.R. 3200, 101st Cong. (1989).
218 The Senate Labor Committee added the subterfuge language to the ADA in its
markup on August 2,1989. See D.C. BAR, DISTRICr OF COLUMBIA BAR TASK FORCE REPORT
ON THE EFFECrS OF THE AMERICANs wrTH DisABIrrIs Acr ON EMPLOYER-SPONSORED
HEALTH PLANS 42,60 n.192 (1993).
2 19
See Older Workers Benefit Protection Act, Pub. L. No. 101-433, 104 Stat. 978 (1990)
(codified in scattered sections of 26 U.S.C.).
220 Farber, supra note 205, at 897.
221 Professors Eskridge and Frickey describe canons of statutory construction as "rules of
thumb," which "enable interpreters to draw inferences from the language, format, and subject
matter of the statute" ESKRIDGE & FRICKEY, supranote 93, at 634.
1572
OHIO STATE L4WJOURNVAL
[Vol. 60:1533
subterfuge was added just before the OWBPA amendments could support an
entirely different inference than the above proposed one. Because the OWBPA
and legislative history speak so clearly about what subterfuge does not mean,
22 2
Congress may not have felt the need to reiterate those sentiments in the ADA,
especially in light of the institutional costs to Congress of amending a statute to
223
override a judicial decision.
b. The Supreme Court on Subterfuge
The Supreme Court in Betts adopted the McMann definition of
subterfuge-- 5 'a scheme, plan, stratagem, or artifice of evasion" 224-which
carries with it a connotation of specific "intent... to evade a statutory
requirement." 225 This approach was explicitly rejected by Congress in the
22 7
OWBPA 226 and in numerous sources of ADA legislative history
Furthermore, even though Betts interpreted the same word- subterfuge-it did
so in the context of the ADEA, not the ADA.22 8 The Court has not yet addressed
222
Representative Owens, after expressing disapproval with the Betts interpretation of
subterfuge, stated that: "Indeed, our committee recently reported out a bill [i.e., the OWBPA]
to overturn the Betts decision for which I voted.... It is not our intent that the restrictive
reading of Betts, with which we do not agree, should be carried over to the ADA." 136 CONG.
REC. H4623 (daily ed. July 12, 1990) (statement of Rep. Owens). Sen. Kennedy made a nearly
identical statement. See supra note 213. These statements suggest that the rejection of Betts in
the OWBPA and in the legislative history were perceived by Congress as sufficient to rule out
such an interpretation in the ADA. Therefore, an ADA amendment was not considered
necessary.
223 Professor Brudney states that: "The initiation, negotiation, and enactment of a statute
is a multidimensional process that requires committing considerable institutional resources,
navigating politically sensitive intemal procedures, and anticipating substantial societal
consequences." Brudney, supranote 132, at 16-17. He then provides a detailed account of the
struggle involved in overturning the Betts decision via the OWBPA. See id.at 17-20. Professor
Brudney goes on to assert that the devotion of time and resources to the OWBPA's judicial
override rendered Congress "marginally less capable' of resolving the meaning of other civil
rights statutes in the same period. See id.at 20.
224 United Airlines, Inc. v. McMann, 434 U.S. 192,203 (1977).
225 Id.
226
See Older Workers Benefit-Protection Act, Pub. L. No. 101-433, 104 Stat. 978 (1990)
(codified in scattered sections of 26 U.S.C.) and supra Part IV.B.2. (discussing congressional
override of Betts).
227
See supraPart IV.C.2.a.
228 Professor Brudney points out some basic differences between the ADA and ADEA
subterfuge provisions. See Brudney, supra note 132, at 100 n.395. One notable difference is in
precise statutory language. The ADA states that safe harbor plans "shall not be used as a
subterfuge to evade... ," while the ADEA exemption (as it appeared to the Betts Court)
JUDICIAL DEFERENCE TO THE EEOC
1573
229
the meaning of subterfuge in the ADA.
Therefore, because Betts was rejected, and the Court has not since addressed
the subterfuge issue under the ADEA or ADA, the lower federal courts should be
looking to sources outside the Supreme Court for guidance on the meaning of
subterfuge. Nonetheless, Betts remains good law in the eyes of many lower
courts.
2 30
c. The EEOCPositionon Subterfuge
According to the EEOC, the word subterfuge "refers to disability-based
disparate treatment that is not justified by the risks or costs associated with the
disability." 23 1 Whether the challenged insurance distinction is being used as a
subterfuge will be determined "on a case by case basis, considering the totality of
the circumstances."2 3 2 The EEOC also provides a non exhaustive list of five
2 33
ways inwhich a respondent can prove that the distinction is not a subterfuge.
applies only to a plan "which is not a subterfuge to evade ... ." Id. The former provision, with
its mandatory language, is obviously more strict.
229 Such an opinion is likely forthcoming. As discussed infra Part IV.C.2.d, a split is
slowly emerging in the appellate courts, and it is only a matter of time before the Supreme
Court will be called upon to decide this contentious issue.
230 See infraPart IV.C.2.d (discussing lower court decisions on subterfuge issue).
231 EEOC Interim Guidance on Health Insurance, supra note 204, at 11. The stated
subject of the guidance is "the application of the Americans with Disabilities Act of 1990 to
disability-based distinctions in employer provided health insurance." Id. at 1. This and other
enforcement guidance is developed by the ADA Policy Division in the Office of the Legal
Counsel for approval by the EEOC Commissioners. For a complete discussion of the
development of EEOC enforcement guidance, see CCR REPORT ON ADA T
I, supranote
1,at 73-76. The Department of Justice has expressed similar views with regard to the meaning
of subterfuge. See Parker v. Metropolitan Life Ins. Co., 121 F.3d 1006, 1012 n.5 (6th Cir.
1997) (discussing, in dicta, the DOJ view on subterfuge in the context of insurance contracts);
Am. With Disabilities Prac. & Compliance Manual (Lawyers Coop) tab 3, at § M1/-3.11000
(1997).
232 EEOC Interim Guidance on Health Insurance, supranote 204, at 11.
2 33 The following justifications are acceptable proof: that the respondent has not engaged
in the disability-based disparate treatment alleged; that the distinction is justified by legitimate
actuarial data or by actual or reasonably anticipated experience; that the distinction is necessary
to satisfy the legally required standards for a fiscally sound insurance plan; that the distinction
is necessary to prevent the occurrence of an unacceptable change either in the coverage of the
health insurance plan or in the premiums charged to maintain it;
or that a disability-specific
treatment which was denied by the plan does not provide any benefit (i.e., has no medical
value). See id.at 11-13.
1574
OHIO STATELA WJOURAAL
[Vol. 60:1533
To further complicate matters, the EEOC issued its subterfuge rule in the
form of interim enforcement guidance. 234 In developing such guidance, the
EEOC has more latitude in deciding what issues to address and when to publish
its guidance, whereas regulations must be issued within a specific time frame
prescribed by Congress. 235 The "interim" or temporary nature of the guidance
allows the agency to refine the rule before casting it in final form. 23 6 Therefore,
the primary advantages of interim enforcement guidance are flexibility and
speed. However, the lack of judicial deference to such guidance seems to offset
these advantages.
Interim enforcement guidance is entitled to the least amount of deference
because it is published on an "interim," not a permanent basis; and because it is
not subject to notice and comment procedure. 237 The interim guidance
addressing the subterfuge issue has a more distinct flaw: it has been "interim"
guidance for nearly six years-since June of 1993.238 In fact, in August of 1997,
the EEOC actually withdrew proposed plans to issue final guidance on
employer-provided health insurance.23 9 Even if a good excuse exists for this
lapse, the agency nonetheless appears sloppy, and thus deprives itself of the
deference that it could secure with final guidance.
2 34
The EEOC issues two other types of guidance which are functionally equivalent to
enforcement guidance: (1) Compliance Manuals, see, e.g., EEOC ComplianceManualSection
902 Definition of the Term "Disability," 8 FEP Manual (BNA) 405:7251 (1995) (discussing
definition of disability); and (2) Technical Assistance Manuals, see, e.g., The Americans with
DisabilitiesAct Title I Technical Assistance Manual: Pre-Employment Inquiries About Drug
andAlcohol Use, Jan. 1992, availablein WL, Database ADA-TAM, File No. I - 8.8.
235 Congress provided that the EEOC issue regulations within one year after the
enactment of the ADA. See 42 U.S.C. § 12116 (1994).
236 The "instructions" in the employer benefits guidance state that "[t]his enforcement
guidance is to be used on an interim basis until the Commission issues final guidance after
publication for notice and comment." EEOC Interim Guidance on Health Insurance, supranote
204, at 1. To date, no such "final guidance" has been issued. The EEOC has used the interim
method in only one other context, but in that case, it did issue final enforcement guidance. See
EEOC Enforcement Guidance on Pre-Employment Inquiries Under the Americans with
Disabilities Act, 8 FEP Manual (BNA) No. 783, at 405:7193 (1995) (former interim
enforcement guidance issued in June of 1991).
237 The "interim guidance" is exempt from notice and comment procedure if, for "good
cause," the agency finds that such procedure is "impracticable, unnecessary, or contrary to the
public interest.' 5 U.S.C. § 553(b)(B) (1994). The EEOC has voluntarily subjected its
"Interpretive Guidance" to APA procedure, even though a specific exemption exists for
interpretive rules in § 553(b)(A) (1994). The agency has not adopted that same strategy for its
enforcement guidance, although arguably, it should.
23 8
See CCR REPORT ON ADA TrrLEI, supranote 1, at 259 (criticizing EEOC for delay
of final guidance).
239
See id.at 147, 259.
19991
JUDICIALDEFERENCETO THE EEOC
1575
d. JudicialResponses to Conflicting Views on Subterfuge
The lower federal courts are split with regard to the meaning and application
of subterfuge. 240 If a court gives significant weight to legislative history, it will
likely conclude that Congress overwhelmingly disapproves of Betts. However,
such a court will remain confused about the precise meaning of subterfuge
because Congress offers no specific guidance about what the word does mean;
even though it speaks quite loudly about what it does not mean. Faced with such
ambiguities, a court may properly defer to the EEOC interim guidance, which
contains the most comprehensive test available to determine the meaning of
subterfuge.
If a court does not trust legislative history, or accepts the timing argument
discussed above,241 it may cite Betts as the controlling authority, or possibly
adopt its own interpretation that conflicts with that of the EEOC. If a court
adopts its own interpretation of subterfuge that is inconsistent with both Betts and
the EEOC, it likely walks a short plank toward reversal. The Supreme Court will
probably do one of two things if and when it addresses the subterfuge issue:
adopt its pro-employer interpretation from Betts, or follow the EEOC position
that favors the disability community. No room exists in this scheme for novel
interpretations of subterfuge.
If a court adopts the Betts interpretation, it must still determine how to apply
it, because neither the ADA nor its legislative history provide any technical
assistance in this matter. The EEOC's enforcement guidance, however, is replete
with such technical assistance. It not only defines the subterfuge rule, but also
describes several situations in which the rule should not apply.242 Despite the
fact that the EEOC's subterfuge rule exists in a form that is not entitled to
controlling weight the EEOC is still a reliable source of expertise in the area of
employment discrimination. 243 Congress provided no helpful language or
context in section 501(c), and the Supreme Court has only defined the word in an
overruled ADEA case. Therefore, the EEOC guidance, at the very least, should
be persuasive.
It is important to note that, according to the majority of the Supreme Court,
EEOC enforcement guidance is not entitled to full Chevron deference because it
240
See infranotes 250-60 and accompanying text (discussing circuit court decisions on
the subterfuge issue).
241 See supraPart IV.C.2.a.
242
See supraPart IV.C.2.c (discussing the EEOC position on subterfuge).
243 Congress has recognized the EEOC's unique expertise in the area of employment
discrimination. See H.R. REP. NO. 92-238 (1971), reprintedin 1972 U.S.C.CA.N. 2137, 2146
("Administrative tribunals are better equipped to handle the complicated issues involved in
employment discrimination cases.").
1576
OHIOSTATE LA WJOURNAL
[Vol. 60:1533
does not constitute a "legislative regulation." 244 However, it does qualify for
earned deference under Skddmore.245 Therefore, the enforcement guidance,
"while not controlling upon the courts by reason of [its] authority, [does]
constitute a body of experience and informed judgment to which courts and
litigants may properly resort for guidance." 24 6 The extension of Skidmore to the
EEOC enforcement guidance is a conservative step,24 7 as some might argue that
this guidance deserves full Chevron deference, even though it is not "legislative"
in character.2 48 Therefore, Skidmore is the least deferential standard that could be
accorded to the EEOC subterfuge rule. Anything less would be no deference at
all.
The Supreme Court has not yet visited the subterfuge issue in the context of
the ADA. It has only commented on the meaning of the word for purposes ofthe
ADEA. Despite Congress's explicit rejection of the Court's position on
249
subterfuge, many appellate and district courts still follow Betts.
244 See supra Part II.B.
245
The Court in Sddmore applied its deference standard to the "rulings, interpretations,
and opinions of the Administrator." Skidmore v. Swift, 323 U.S. 134, 140 (1994). EEOC
enforcement guidance would certainly fall within this broad category of agency guidance to
which the Skidmore standard applies.
246
Id. The traditional Slddmore factors used to determine the persuasiveness of agency
guidance, such as "thoroughness" and "validity," require largely normative analysis. One may
argue that EEOC enforcement guidance is thorough, by revealing agency fact finding; or valid,
by presenting policy arguments in favor of the EEOC position. However, it may be too
argumentative, or simply futile, for this Note to speculate about the validity or thoroughness of
agency guidance, especially without any judicial authority to support such speculation. These
arguments are likely more appropriate for litigants than law students.
247 As Professor Colin Diver points out, Sddinore deference "might mean nothing more
than 'respectful or courteous regard' for the agency rule. See Diver, supranote 9, at 565.
248 Justice Antonin Scalia has argued that non legislative EEOC guidance deserves full
Chevron deference. See EEOC v. Arabian Am. Oil Co., 499 U.S. 244, 260 (1991) (Scalia, J.,
concurring) (arguing that EEOC interpretation "need only be reasonable to be entitled to
deference") (quoting EEOC v. Commercial Office Prods. Co., 486 U.S. 107, 115 (1998)). One
commentator has suggested that the Slddmore factors are applied in the second prong of
Chevron to determine whether an agency construction is "reasonable." See Yavelberg, supra
note 11, at 183 n.96. Whether or not this is a proper interpretation of Skidmore or Chevron is
beyond the scope of this Note, but well within the scope of another. See id. at 183-84. For
purposes of this discussion, Slddmore is accorded its traditional interpretation, under which a
court pays courteous regard to non legislative agency rules, but does not apply Chevron
deference.
24 9
It is not quite clear why lower courts still adhere to Betts. It is possible that they have
no predisposition toward EEOC guidance and no awareness of the second-class status of the
EEOC's guidance. They may simply be more comfortable diving into Supreme Court case law,
rather than into the mire of administrative law.
1999]
JUDICIAL DEFERENCETO THE EEOC
1577
The District of Columbia Circuit adopted the Betts definition in Modderno v.
King.250 Plaintiff, a beneficiary of a Foreign Service Benefit Plan, brought an
51
action alleging discrimination under section 504 of the Rehabilitation Act
based on her mental disability because her plan imposed a $75,000 lifetime
maximum for mental health benefits. No such cap existed for physical disability
benefits. In determining whether this cap fell within the safe harbor provision,
the court followed Betts, which imposes a specific intent to evade as a
prerequisite to a finding of subterfuge. Because the challenged coverage
limitations were adopted in 1990, before ADA standards became applicable to
section 504, "the limitations in the Plan cannot constitute a subterfuge to evade
' 252
congressional purposes.
The court; citing Betts, brushed off the EEOC interpretation with apparent
ease. "[E]ven assuming deference to the agency under [Chevron], the costjustification requirement was 'at odds with the plain language of the statute
itself.' 253 This is risky reasoning because, first, Congress expressly renounced
the Betts definition in legislative history, a fact that is not discussed in the
opinion. Second, this reasoning is incomplete because the court does not point
out how or why the EEOC view conflicts with the ADA. It simply leaps to the
conclusion reached in Betts.
The same result was reached in Krauel v. Iowa Methodist Medical
Center.2 54 With regard to EEOC guidance, the Eighth Circuit simply agreed with
255
the "conclusions reached in Modderno and adopt[ed] them as [their] own."
25 6
The court also disregarded numerous sources of legislative history, stating that
if Congress had intended to reject Betts, it should have done so by removing
2 57
subterfuge from the ADA.
In a more recent case addressing the subterfuge issue, Ford v. SheringPlough Corp.,25 8 the Third Circuit adopted the Betts interpretation. However, the
250 82 F.3d 1059, 1064 (D.C. Cir. 1996).
251 See 29 U.S.C. § 794 (1994).
2 52
Modderno, 82 F.3d at 1064. The court brushed off the notion that the OWBPA
prevents the application of Betts to the ADA. It adopted the popular argument that, because
Congress was aware of the Betts interpretation when it adopted the subterfuge language, it
acquiesced to the Supreme Court interpretation. See id. at 1064-65. This argument is discussed
further above. See supraPart IV.C.2.a.
2 53
Moddenzo, 82 F.3d at 1065 (quoting Public Employees Retirement Sys. v. Betts, 492
U.S. 158, 171 (1989)).
254 95 F.3d 674 (8th Cir. 1996).
255 Id at 679.
25 6 See supraPart IV.C.2.a (citing and discussing legislative history of subterfuge rule).
2 57 See Krauel,95 F.3d at 679.
258 145 F.3d 601 (3d Cir. 1998).
1578
OHIO STATE LAWJOURFAL
[Vol. 60:1533
court conducted its entire subterfuge analysis without any reference to the EEOC
guidance. It is possible that the EEOC view was not briefed by the parties in
Ford, considering the past failures of the rule. It is also possible that upon
examining prior appellate precedent rejecting the rule, the court believed that the
EEOC view was not worth mentioning. Either way, it is quite surprising that the
agency charged with enforcement of the ADA is completely ignored in a case
where it has promulgated a rule directly on point. This may illustrate the EEOC's
second-class status even more than cases that explicitly reject its guidance.
Because no other appellate courts have directly addressed the ADA
subterfuge issue,2 59 the EEOC is "batting zero" in the circuit courts with respect
to its definition of subterfuge. 2 60 The most powerful illustration of this lack of
deference is seen in Ford,in which the Third Circuit examines Betts and the
legislative history which denounces it without even mentioning EEOC guidance.
Therefore, the EEOC has not even served the modest role of tiebreaker between
competing interpretations of the ADA safe harbor provisions.
Having examined the overall status of the EEOC in the courts, two questions
remain. First why does the EEOC receive less deference than other agencies?
Second, how can the EEOC restore its credibility in the courts?
V. REASONS FOR A LACK OF DEFERENCE TO THE EEOC
One reason for the present lack of deference to the EEOC could be the
agency's tumultuous history in the courts. The EEOC arose from a highly
controversial statute (Title VII), was given weak enforcement power under that
statute, and received little deference in its early interpretations. This history may
have cast a shadow over EEOC guidance that has drifted into the ADA context
even though the ADA delegates much greater enforcement power to the EEOC.
Therefore, because of the failures of the agency in the past courts may attach a
mild presumption of invalidity to EEOC guidance, and ultimately strike down
that guidance in a reflexive manner.
259 The Sixth Circuit in Parkerv. MetropolitanLife InsuranceCo., did provide a detailed
explanation of the subterfuge issue in a footnote. See 121 F.3d 1006, 1013 n.7 (6th Cir. 1997).
However, the court left the resolution of the issue "for another day" concluding that the
challenged disability plan was not covered by ADA Title II (the public accommodations
provision), and therefore, "the safe harbor provision [was] not implicated."IaJ
2 60 The district court record was not sufficiently developed to warrant an empirical
analysis. I found approximately 34 district court cases that directly addressed the ADA
subterfuge issue. However, many of those cases did not articulate their position on deference to
the EEOC interim guidance. A handful of district courts applied the Betts interpretation, but did
not go so far as to reject the EEOC interpretation. In essence, the district court record presents a
muddled view of the subterfuge issue. Considering the weak and temporary nature of district
court authority, it was simply not worth it to sift through the mud.
1999]
JUDICIALDEFERENCE TO THE EEOC
1579
A second reason for the lack of deference may be the non technical nature of
the subject matter over which the EEOC exerts its regulatory power. In contrast,
other agencies promulgate rules and regulations in highly complex or scientific
areas of the law. For example, determining the precise amount of benzene
allowed in drinking water or the types of economic data required for a valid
prospectus requires decisions that ring with agency expertise. The EEOC's
domain, employment discrimination, is not an area that requires such technical
expertise; rather, logic and basic fairness concerns may dominate
decisionmaking in this area. Simply put, judges believe that they do not need an
agency telling them who has or has not suffered wrongful discrimination; yet
they welcome agency guidance in technical areas where judicial reasoning is not
2 61
sufficient.
A third reason arises from the actions of Congress itself. Congress often
writes ambiguous statutes when dealing with controversial issues. The idea is
that agencies will fill any gaps left by Congress, and Congress thereby escapes
accountability. 262 The problem is that this allows large amounts of power to shift
to agencies which are not directly accountable to the public. When the EEOC is
called upon to interpret ambiguous statutory terms such as substantial limitation,
or subterfuge, courts are naturally skeptical of its conclusions because,
presumably, Congress should be making these sensitive policy decisions. By
contrast, an agency interpreting its own regulations, rather than ambiguous
statutory terms, is generally given more deference. 263 Therefore, the EEOC
takes on a distinct burden when it is called upon to interpret ambiguous statutory
terms, especially when such interpretations result in profound policy decisions.
261 The D.C. Circuit recognizes this reality. It stated that, when dealing with "a highly
technical question... [,] courts necessarily must show considerable deference to an agency's
expertise." MCI Cellular Tel. Co. v. FCC, 738 F.2d 1322, 1333 (D.C. Cir. 1984); see also
Appalachian Power Co. v. EPA, 135 F.3d 791,797 (D.C. Cir. 1998) (upholding an EPA rule
with cognizance of the "deference due to an agency dealing with largely scientific and
technical matters").
262
A cynical description of this drafting phenomenon is "punting?' One commentator
described the phenomenon as follows: "In the face of uncertainty and factional pressures,
Congress can simply punt, by rather vaguely identifying whatever problem is pressed upon it,
and then passing it on to an agency." William T. Mayton, The Illegitimacy of the Public
Interest Standardat the FCC,38 EMORY LJ. 715, 728 (1989).
263 This view was expressed by the D.C. Circuit in Birt v. Surface TransportationBoard,
90 F.3d 580 (D.C. Cir. 1996). Discussing deference to a rule issued by the Interstate
Commerce Commission, the court stated that "precedent firmly establishes that courts owe
'even greater deference to agency interpretations of agency rules than ... to agency
interpretations of ambiguous statutory terms."' Id. at 589 (quoting Capital Network Sys., Inc. v.
FCC, 28 F.3d 201,206 (D.C. Cir. 1994)).
1580
OHIOSTATELAWJOURNAL
[Vol. 60:1533
A fourth reason why the EEOC has received little deference arises from its
own carelessness. 264 The interim guidance containing the subterfuge rule is a
prime example. This rule has existed in "interim" form for nearly six years,
during which a split has arisen in the courts as to the meaning of the rule. The
result is a confusing and unpredictable standard for application of the ADA safe
harbor provision. The EEOC could eliminate this result quite easily by either
repealing the interim guidance or subjecting it to the procedures necessary to
establish a final rule. Either way, the agency would be acting responsibly to
communicate its final position to the public. In the interim, litigants and courts
will continue to struggle with the meaning of subterfuge.
A fifth possible reason exists for the lack of deference to the EEOC. It arises
from the internal criticisms levied at the EEOC from bureaucratic entities such as
the United States Commission on Civil Rights (CCR). For example, Mary
Frances Berry, Chairperson of the CCR, made the following comment with
regard to the EEOC: "In general, the Commission found that [the] EEOC is not
fully responsive to the needs and views of its stakeholders, including individuals
'265
with disabilities, employers, disability professionals, and disability experts.
This statement appeared in an elaborate report issued by the CCR in September
of 1998, assessing the EEOC's enforcement efforts in the context of Title I of the
ADA. The Commission's criticism ranged from shoddy drafting to overuse of
non-legislative rules. When an entity such as the CCR devotes a considerable
amount of resources to the topic of EEOC effectiveness, and publishes an entire
report on that very topic, one can safely assume that the EEOC's effectiveness is
questionable. When an agency is the target of such powerful criticism by its
political peers, that message could ultimately affect the EEOC's credibility
within the courts.
264
Criticizing the EEOC as a careless agency may be too harsh, considering the
"seemingly insurmountable problems" encountered by the agency. CCR REPORT ON ADA
TITLE I, supranote 1, at 51. Aside from the weak enforcement powers originally granted to the
EEOC, the agency has suffered from numerous management turnovers, insufficient staff, lack
of training, and an enormous backlog of cases. See UNITED STATES COMM'N ON CIVIL RIGHTS,
FEDERAL CIVIL RIGHTS ENFORCEMENT EFFORT 86-88 (1971). Furthermore, the enactment of
the ADA in 1990 and the Civil Rights Act of 1991 resulted in a 26% increase in the number of
adjudicatory charges filed. See Oversight Hearing on the Equal Employment Opportunity
Commission: HearingsBefore the Subcomm. on Select Educ. and Civil Rights of the House
Comm. on Educ. and Labor, 103d Cong. 6 (1993) (statement of Linda G. Morra, Director,
Education and Employment Issues, Human Resources Division General Accounting Office).
265
See CCR REPORT ON ADA TITLE I, supranote 1, at iii.
1999]
JUDICIALDEFERENCETO THEEEOC
1581
VI. RECOMMENDATIONS
Having established the EEOC's status as a second-class agency, and some
possible reasons for that status, this Note focuses on the means of redeeming the
EEOC from its inferior status. The following analysis targets problems
associated with particular EEOC rules and guidelines (some of which were not
discussed extensively above), and offers recommendations specifically tailored
towards those problems. Here, the major concern is not just increased deference
or credibility within the courts, but overall improvement of agency functioning.
The first issue is clarity. The EEOC's enforcement guidance on psychiatric
disabilities provides one example in which lack of clarity in EEOC guidance can
have unfortunate consequences. 26 6 Among other things, the guidance states that
the DSM-1T4 67 may be "relevant" to identifying mental disorders that qualify for
ADA coverage. 268 The guidance states that not all conditions in the DSM-IV are
disabilities for purposes of the ADA.269 What is unclear from this guidance is
whether the DSM-IV is the only relevant source. The EEOC does not state its
270
position on conditions set forth in other texts used to classify mental disorders
or on conditions that are not documented in any authoritative text. This has lead
to concerns in the psychiatric community that the EEOC enforcement guidance
may be encouraging a proliferation of frivolous ADA claims based on mental
266
See generally Enforcement Guidance: The Americans with Disabilities Act and
Psychiatric Disabilities, EEOC Compl. Man. (BNA) No. 915.002 (Mar. 25, 1997) [hereinafter
EEOC Guidance on Psychiatric Disabilities]. This document was intended to facilitate
enforcement of the ADA on charges alleging employment discrimination based on psychiatric
disabilities (which fall under the "mental impairment" prong of the ADA definition of
disability, 42 U.S.C. § 12102(2) (1994)). For a complete discussion of the EEOC guidance on
psychiatric disabilities, see generally Sarah Starnes, PsychiatricDisabilitiesand the ADA: An
Analysis ofConventionalDefensesandEEOC Guidelines, 18 REV. LMG. 181 (1999).
26 7
AMRICAN PSYCHIATRIC AWS'N, DIAGNOsTIC AND STATISTICAL MANUAL OF MENTAL
DISORDERS (4th ed. 1994). "mhe DSM-IV has been recognized as an important reference by
the courts and is used widely by American mental health professionals for diagnostic and
insurance reimbursement purposes." CCR REPORT ON ADA TT= I, supranote 1, at 119.
26 8 See EEOC GUIDANCE ON PSYCHIATRIC DISABILrrMEs, supranote 266, at 3.
269 EEOC regulations state that the following are not mental disabilities: transvestitism,
transsexualism, pedophilia, exhibitionism, voyeurism, gender identity disorders not resulting
from physical impairments, or other sexual behavior disorders, compulsive gambling,
kleptomania, pyromania, psychoactive substance use disorders resulting from current illegal
use of drugs, homosexuality, and bisexuality. See 29 C.F.R § 1630.3(d)-(e) (1997).
270
Two "other valuable tests used to classify mental disorders [are] the International
ClassificationofDiseases,and the Guides to the EvaluationofPermanentImpairment,Fourth
Edition." CCR REPORT ON ADA TM.LI, supranote 1, at 120.
1582
OHIOSTATELAWJOURNAL
[Vol. 60:1533
illnesses not covered in the DSM-IV or any other text.2 71 Thus, one response to
the above problem is for the EEOC to clarify whether the DSM-1V is the only
relevant source for determining mental disabilities, and if not, then the EEOC
should identify all of the "relevant' diagnostic texts. 272 As a general
recommendation, the EEOC should clarify known ambiguities in its guidance.
Lack of clarity is often an unavoidable result of any drafting effort.
However, the EEOC, unlike Congress, can quickly and easily resolve
ambiguities arising from its prior pronouncements simply by issuing some form
of expedited policy guidance.2 73 Therefore, if the EEOC is aware of any serious
interpretive problems with its guidance (and entities such as the CCR certainly
facilitate that awareness), it is difficult to excuse the agency's inaction.
Another ground for criticism of EEOC guidance is inconsistency.2 74 The
EEOC guidance on employer-provided health benefits provides helpful
illustration. The EEOC states clearly that it does not consider distinctions
between mental and physical health benefits to be disability-based
distinctions. 275 "However, EEOC does consider this precise distinction to be a
disability-based distinction in the context of long term or disability
insurance." 276 As one source has noted, the EEOC's distinction between
"mental" and "physical" for health insurance but not for disability insurance
"abrogates the important purpose of the ADA to prevent any kind of different
271 See id.; see also James J. McDonald, Jr. & Jonathan P. Rosman, EEOC Guidance on
PsychiatricDisabilities:Many Problems, Few Workable Solutions, in NATIONAL SYMPOSIUM
ON PsYcHIATRIc DIsABILrrlEs AND THE EEOC's NEW ENFORCEMENT GUIDANCE UNDER THE
AMERICANs wrrH DIsABII
272
SAcr 8 (Jan. 12-13 1998).
See CCR REPORT ON ADA TrLE I, supranote 1, at 120.
273 Policy guidance, like interpretive guidance and enforcement guidance, is exempt from
the procedural requirements of the APA. See 5 U.S.C. § 553(b)(A) (1994).
274
Recall that consistency of earlier and later pronouncements is one element of the
Skidmore test. See supra Part I.A. Therefore, a lack of consistency could be detrimental to an
agency upon judicial review of its guidelines.
275
See EEOC Interim Guidance on Health Insurance, supranote 204, at 6. Whenever it
is alleged that a health-related term or provision of an employer-provided health insurance plan
violates the ADA, the first issue is whether the term or provision is a disability-based
distinction. See id. at 3. With regard to mental and physical benefits, the EEOC states:
Typically, a lower level of benefits is provided for the treatment of mental/nervous
conditions than is provided for the treatment of physical conditions.... [A]Ithough such
distinctions may have a greater impact on certain individuals with disabilities, they do not
intentionally discriminate on the basis of disability and do not violate the ADA.
Id. at 6.
27 6
CCR REPORT ON ADA TTLEI, supranote 1,at 259.
JUDICIALDEFERENCETO THE EEOC
1999]
1583
treatment on the basis of disability." 2 77 This confusion as to the EEOC's position
supports the need for a final version of enforcement guidance, in which the
EEOC could simply acknowledge or actively correct its own inconsistent
pronouncements.
Probably the most sweeping criticism of current EEOC rulemaking practices
is its excessive use of non binding forms of guidance. The EEOC has clear
rulemaking authority over the ADA, and can, therefore, issue rules with the force
of law upon satisfaction of certain procedures. The agency in recent years has
nonetheless favored non binding forms of guidance, such as interpretive
guidance, enforcement guidance, and technical assistance manuals. This shift
reflects valid concerns on the part of the agency to avoid the heavy burdens of
administrative procedure and to effectuate its policies more quickly.27 8 However,
the EEOC's departure from traditional rulemaking effectively excludes
279
stakeholders from the policy development process.
With regard to public participation in the development of non legislative
EEOC rules, the CCR states that the 'EEOC 'floats' ideas at meetings and
welcomes letters on policy issues from stakeholders, but it does not solicit
comments on or circulate drafts of proposed policies outside of the agency before
they are approved by the Commissioners."2 80 In response to this problem, the
CCR recommends the creation of an ADA Policy Advisory Committee,
consisting of representatives from interested groups, to ensure interaction with
stakeholders.2 81 It also recommends a system of notice in the FederalRegister
282
that apprises stakeholders of EEOC policy developments.
The above recommendations of the CCR are probably valuable mechanisms
for ensuring public participation and agency accountability; however, a much
simpler solution exists. The EEOC could simply revive the APA notice and
comment procedure that has been firmly established in the courts. This may be
277 Id.
278
APA notice and comment procedure is not the only hoop through which the EEOC
must jump. Like other agencies, the EEOC must also navigate the requirements of the
Regulatory Flexibility Act (RFA), 5 U.S.C. §§ 601-612 (1994), the Contract with America
Advancement Act, 5 U.S.C. §§ 801-808 (Supp. 111 1997), and Exec. Order No. 12,866, 3
C.F.R. 638 (1994), reprintedin 5 U.S.C. § 601 at 557-61 (1994), issued by President William
J. Clinton. These documents impose numerous procedural requirements above and beyond
those of the APA, which affect every stage of the rulemaking process.
279 It is quite obvious that a primary purpose of the APA notice and comment procedure
embodied in 5 U.S.C. § 553 (1994) is to allow for public participation in the rulemaldng
process.
2 80
CCR REPORT ON ADA TrILE I, supranote 1, at 248.
281 See id.
2 82
See id.
1584
OHIO STATE LA WJOURNAL
[Vol. 60:1533
even more burdensome to the agency than the combined recommendations of the
CCI. However, what the EEOC loses in efficiency, it would almost certainly
gain in credibility; that is, courts are much more willing to defer to agency
guidance that has satisfied the requirements of the APA.2 83 Therefore, the most
effective solution is not a radical one. Absent special circumstances, 2 84 the
EEOC should rely primarily upon notice and comment rulemaking when
2 85
drafting controversial ADA guidance.
VII. CONCLUSION
The EEOC can fairly be characterized as a second-class agency. The
grounds for this conclusion lie in an empirical analysis of Supreme Court cases,
which yielded a seventy-two percent deference rate for all agencies, compared to
a fifty-four percent rate of deference to the EEOC. This conclusion is further
reinforced by an individual analysis of the Court's deference cases under each of
the statutes that the EEOC administers.
When examining lower court deference cases under the ADA, a statute that
contains an impeccable grant of rulemaking authority to the EEOC, the agency's
inferior status was further reinforced. Lower courts addressing two EEOC
rules-the mitigating measures rule and the subterfuge rule-were reluctant to
grant any deference to the EEOC. As the mitigating measures rule has
progressed in the judicial system, judges interpreting that rule seem to have taken
a more deferential stance to the EEOC. With-regard to the subterfuge rule, quite
the opposite has occurred. Every appellate court to date has rejected the EEOC's
283 The Supreme Court in GeneralElectric Co. v. Gilbert, 429 U.S. 125, 141 (1976),
U.S. 1079 (1977), acknowledged the cleansing effect of APA notice and
comment procedure on EEOC guidance in the Title VII context See supra Part .IIA.2.
reh'g denied, 429
2 84
Numerous instances arise where the EEOC is purely justified in ignoring APA
procedures, and few courts would challenge the agency's decision. Examples include situations
in which the agency effectuates technical and non controversial changes in its guidance and in
which it reacts to an emergency situation affecting public health or safety. See LUBBERS, supra
note 61, at 78-83.
285 There is a middle ground between notice and comment rulemaking and rulemaling
that is purely exempt from APA procedure. This method is "direct-final" rulemaking. See id, at
84-87. In direct rulemaking, an agency publishes its proposed rule in the FederalRegister,
with a caption stating that the rule will take effect as written within thirty days unless it receives
adverse comment. If such comments appear, then the agency will commence the traditional
APA procedures for notice and comment rulemaking. If not, then the rule is published as a
final, binding rule. The problem with this method in the EEOC context, especially with ADA
guidance, is that the agency will almost certainly receive negative commentary. Considering
the strikingly divergent interests between the disability community and employers (especially
small businesses), it would be difficult for the EEOC to draft a proposed rule that pleases
everyone.
1999]
JUDICIALDEFERENCE TO THEEEOC
1585
construction of the word subterfuge, and the Third Circuit in Fordv. SheringPlough Corp.286 simply ignored the EEOC rule in its subterfuge analysis. A case
such as Fordin which the EEOC view was entirely ignored, even though it had
promulgated guidelines directly on point, is probably the most striking
illustration of the EEOC's second-class status.
One reason for the EEOC's inferior status could be the non technical nature
of its guidelines because courts are much more willing to defer to an agency
whose expertise is in complex, scientific, or highly technical areas. Another
reason can be traced to Congress's tendency to leave gaps in its statutes by way
of ambiguous terms, with the knowledge (or at least hope) that an agency will fill
those gaps. Although this phenomenon is common among many statutes, and is
dealt with by many agencies, it is possible that the controversial nature of
employment discrimination statutes causes Congress to adopt more ambiguous
language, and thereby compels more quasi-legislative activity on the part of the
EEOC. Finally, the EEOC's status can be attributed to its own carelessness. The
agency has engaged in a dangerous pattern of issuing non binding rules,
presumably to avoid the practical burdens of APA procedure. Courts offer less
deference to agency guidelines that have not been cleansed by APA procedures,
even if such guidelines are legitimately exempted from the requirements of the
APA. EEOC guidelines have also been criticized for their lack of clarity and for
their inconsistencies with other pronouncements.
Despite all the criticism levied against the EEOC, it still has a chance to
redeem itself from second-class status. It may do so by closely observing APA
procedure or by issuing more authoritative sources of guidance. However, to
gain back its respect as an administrative agency, the EEOC may simply have to
draft more conservative guidelines. Many courts and citizens feel that the EEOC
has gone too far, that it is attempting to legislate rather than regulate. No amount
of procedure can remedy this underlying distrust of the EEOC. Unless the EEOC
restores that trust,its written guidance may never receive proper deference.
286 145 F.3d 601 (3d Cir. 1998).
1586
OHIOSTATE LAWJOURNVAL
[Vol. 60:1533
APPENDICES
Appendix A. EEOC View Accepted
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Faragher v. City ofBoca Raton, 118 S.CL 2275,2292(1998).
Bragdonv.Abbott, 118 S. Ct. 2196,2209 (1998).
International Union v. Johnson Controls, Inc., 499 U.S. 187, 200 (1991).
Price Waterhouse v. Hopkins, 490 U.S. 228, 253-54 (1989).
EEOC v. Commercial Office Products Co., 486 U.S. 107, 115-16(1988).
Johnson v. Transportation Agency, 480 U.S. 616, 643 n.2 (1987) (Stevens, J.,
concurring).
Western Airlines v. Criswell, 472 U.S. 400,417 (1985).
Washington County v. Gunther, 452 U.S. 161, 190 (1981).
EEOC v. Associated Dried Goods Corp., 449 U.S. 590, 600 n.17 (1981).
Oscar Mayer & Co. v. Evans, 441 U.S. 750,761 (1979).
Nashville Gas Co. v. Satty, 434 U.S.136, 142 n.4 (1977).
Tram World Airlines, Inc., v. Hardison, 432 U.S. 63, 77 (1977).
McDonald v. Santa Fe Trail Transportation Co., 427 U.S. 273, 279-80 (1976).
Albermarle Paper Co. v. Moody, 422 U.S. 405,431 (1975).
Griggs v. Duke Power Co., 401 U.S. 424,433-34 (1971).
Appendix B. EEOC View Rejected
1.
2.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Sutton v. United Airlines, Inc., 119 S. Ct 2139,2146-47 (1999).
Walters v. Metropolitan Education Enterprises, Inc., 519 U.S. 202, 207 (1997).
EEOC v. Arabian American Oil Co., 499 U.S. 244,257-58 (1991).
Public Employees Retirement System of Ohio v. Betts, 492 U.S. 158, 171-72
(1989).
Ansonia Board of Education v. Philbrook, 479 U.S. 60, 69 n.6 (1986).
American Tobacco Co. v. Patterson, 456 U.S. 63, 70 (1982).
Mohasco Corp. v. Silver, 447 U.S. 807, 825 (1980).
City of Los Angeles, Deptartment of Water & Power v. Manhart, 435 U.S. 702,
719-20 (1978).
International Board of Teamsters v. United States, 431 U.S. 324, 378-80 (1977)
(Marshall, J., dissenting).
General Electric Co. v. Gilbert, 429 U.S. 125, 140-43 (1976).
Washington v. Davis, 426 U.S. 229, 263 (1976) (Brenn, J., dissenting).
Espinoza v. Farah Manufacturing Co., 414 U.S. 86,94 (1973).
Phillips v. Martin Marietta Corp., 400 U.S. 542, 545-46 (1971) (Marshall, J.,
concurring).
1999]
JUDICIAL DEFERENCETO THE EEOC
1587
Appendix C. MitigatingMeasures Rule Accepted
1. Bartlett v. New York State Board of Law Examiners, 156 F.3d 321, 329 (2d Cir.
1998).
2. Washington v. HCA Health Services of Texas, Inc., 152 F.3d 464, 469-70 (5th
Cir. 1998).
3. Baert v. Euclid Beverage, Ltd., 149 F.3d 626,629 (7th Cir. 1998).
4. Hamilton v. Southwestern Bell Telephone Co., 136 F.3d 1047, 1051 n.13 (5th
Cir. 1998).
5. Arnold v. United Parcel Service, Inc., 136 F.3d 854,863 (lst Cir. 1998).
6. Matzak v. Frankford Candy & Chocolate Co., 136 F.3d 933, 937 (3d Cir. 1997).
7. Burch v. Coca-Cola Co., 119 F.3d 305,317 (5th Cir. 1997).
8. Foreman v. Babcock & Wilcox Co., 117 F.3d 800, 805 (5th Cir. 1997).
9. Doane v. City of Omaha, 115 F.3d 624,627(8th Cir. 1997).
10. Harris v. H & W Contracting Co., 102 F.3d 516, 520 (1 lth Cir. 1996).
11. Holihan v. Lucky Stores, Inc., 87 F.3d 362,366 (9th Cir. 1996).
12. Daugherty v. City of El Paso, 56 F.3d 695, 696 n.1 (5th Cir. 1995).
13. Roth v. Lutheran General Hospital, 57 F.3d 1446, 1454 (7th Cir. 1995).
Appendix D. MitigatingMeasuresRule Rejected
1. Sutton v. United Airlines, Inc., 130 F.3d 893, 902 (10th Cir. 1997).
2. Gilday v. Mecosta County, 124 F.3d 760, 766 (6th Cir. 1997) (Guy, J.,
concurring).
3. Ellison v. Software Spectrum, Inc., 8-5 F.3d. 187, 191 n.3 (5th Cir. 1996).